Schizophrenia : auditory hallucinations may be established by infancy


Some people suffering from severe mental illness, particularly schizophrenia, hear “voices,” known as auditory hallucinations.

This symptom, which afflicts more than 80% of patients, is among the most prevalent and distressing symptoms of schizophrenia.

Patients “hear voices” speaking to them or about them without anyone actually being there.

Auditory hallucinations, which usually begin in adolescence and young adulthood, “sound” very real to patients and can have a devastating impact on their quality of life because the “voices” are typically distressing and distracting, sometimes compelling the sufferer into suicidal or violent actions.

Uncovering the biological origins of auditory hallucinations is essential for reducing their contribution to the disease burden of schizophrenia.

To investigate the biological origins of hearing “voices” in patients with schizophrenia, a team led by researchers at the Icahn School of Medicine at Mount Sinai used ultra-high field imaging to compare the auditory cortex of schizophrenic patients with healthy individuals.

They found that schizophrenic patients who experienced auditory hallucinations had abnormal tonotopic organization of the auditory cortex.

Tonotopy is the ordered representation of sound frequency in the auditory cortex, which is established in utero and infancy and which does not rely on higher-order cognitive operations.

The study findings, which appears this week in the Nature Partner Journal NPJ Schizophrenia, suggest that the vulnerability to develop “voices” is probably established many years before symptoms begin.

“Since auditory hallucinations feel like real voices, we wanted to test whether patients with such experiences have abnormalities in the auditory cortex, which is the part of the brain that processes real sounds from the external environment,” says Sophia Frangou, MD, PhD, Professor of Psychiatry at the Icahn School of Medicine at Mount Sinai. ”

Specifically, the research team used an ultra-high field scanner with a powerful 7 Tesla magnet to obtain high-resolution images of brain activity while study participants listened passively to tones across a range of very low to very high frequencies.

In healthy brains, these sounds are processed in a very organized fashion; each frequency activates a specific part of the auditory cortex forming a tonotopic map.

The team obtained tonotopic maps from 16 patients with schizophrenia with a history of recurrent auditory hallucination and 22 healthy study participants.

They found that patients showed greater activation in response to most sound frequencies.

Additionally, the mapping of most sound frequencies to parts of the auditory cortex appeared “scrambled” in patients with schizophrenia, suggesting that the normal processes for the organized representation of sound in the brain are disrupted in schizophrenia.

“Because the tonotopic map is established when people are still infants and remains stable throughout life, our study findings suggest that the vulnerability to develop “voices” is linked a deviance in the organization of the auditory system that occurs during infancy and precedes speech development and the onset of psychotic symptoms by many years.

This is particularly exciting because it means that it might be possible to identify potential vulnerable individuals, such as the offspring of schizophrenia patients, very early on.”

Vulnerability to develop “voices” is probably established many years before symptoms begin.

According to the authors, in addition to helping doctors spot people who are likely to experience hallucinations before the symptoms appear or become severe, the auditory cortex may be an area of consideration for novel neurmodulation methods to help patients who already have symptoms.

Looking ahead, Dr. Frangou’s research team will replicate and expand the current observations in larger samples to determine their relevance to hallucinations across diagnoses and to quantify the association of tonotopic disruption to auditory cortical activation and connectivity during actual hallucinatory experiences.

Funding: The study was supported by the National Institutes of Mental Health, the National Cancer Institute, The Netherlands Organisation for Health Research and Development, the Stanley Foundation and the Brain and Behavior Research Foundation.

Recent trends in psychiatric research and treatment initiatives have focused on specific symptoms or experiential features of psychiatric syndromes, while de-emphasizing conventional diagnostic or classification frameworks.

This includes the National Institute of Mental Health’s Research Domain Criteria, a strategic initiative that encourages the adoption of a trans-diagnostic matrix of units of analysis (ie, from molecules to self-report) regardless of the syndromic/diagnostic context.1 In addition, research groups like the International Consortium on Hallucination Research (ICHR) have encouraged new projects that focus on hallucinations across a range of physical and psychiatric conditions as well as nonclinical populations.2 

These and other initiatives have the advantage of supporting high-quality research on segmented aspects of human experience, deliberately avoiding the assumption that diagnostic syndromes represent the most valid and useful object of enquiry.3 However, such a focus risks losing sight of the complex and dynamic experiential context(s) in which hallucinations (or any construct) may arise, which might be essential for determining etiologies, treatments, and other clinical concerns.4,5

In parallel, a limitation of much research in this area is the dominance of operationalized definitions and measurement tools to maximize reliability and simplify diagnostic criteria. While this emphasis on operationalization may help to achieve reliable and interpretable data, it risks a premature simplification of psychopathology.6 

This has prioritized the most easily operationalized symptoms such as hallucinations and delusions over those that involve more individually variable or less easily communicated subjective experiences.7,8 In addition, what counts as a “symptom” is impacted by cultural paradigms and expressions; Western medical models, and perhaps the language of Western society in general, may not supply the words or concepts for people to relate subtle and fluctuating forms of experience, potentially restricting the psychiatric phenomena that are reported by patients or attended to by clinicians.9,10

 Hallucinations are especially prone to these issues due to the apparent simplicity of their mainstream definition.

Their seemingly straightforward characterization belies an experiential complexity that, while not new to literature on these phenomena, is currently borne out by many strands of contemporary research.11–14 Indeed, as Berrios and Marková15 note, the conceptual history of hallucinations is a long and complex one, and includes debates about the representational vs nonrepresentational nature of hallucinations, their equivalency with sensory perception, and the distinction (if any) between hallucinations found in psychiatric vs neurological or other organic conditions.

This article aims to broaden this lens by examining key areas of experience that could contextualize hallucinations within the larger realm of mental or experiential states associated with psychosis or psychotic vulnerability: cognitive experience, perceptual experience, selfhood and sense of reality, temporality, interpersonal experience, and embodiment (supplementary material note i). In doing so, we hope to be able to capture the vast heterogeneity within diagnostic categories, especially the diagnosis of schizophrenia (supplementary material note ii). We then discuss phenomenological and neurocognitive models that attempt to integrate these diverse transformations (supplementary material note iii). Finally, we consider contextual factors that may impact experience, especially the potential role of trauma in the development of psychosis. In doing so, we hope to identify numerous areas for future research that would assist in promoting more effective and valid approaches to assessment, study design, and intervention.Go to:

Domains of Experience

Cognitive Experience and Stream of Consciousness

Cognition in schizophrenia is typically discussed in empirical literature with respect to changes in memory, attention, and executive functioning, which are believed to underlie disorganized thought and behavior (eg, derailment, tangentiality). However, the way in which thought, memory, and attention are experienced is also notably altered in schizophrenia. These include features usually associated with acute psychosis such as thought withdrawal and insertion, thought broadcasting, and passivity experiences. They also include more subtle experiences, such as loss of thought ipseity (the sense that thoughts do not belong to oneself) and spatialization of experience (the sense that thought has spatial or dimensional qualities).16 Such transformations of cognitive experience may be present in all stages of schizophrenia (including the prodrome, acute psychosis, and more long-term or chronic stages)17 as well as in the nonpsychotic tails of its spectrum conditions (eg, schizotypal personality disorder, schizophrenia in remission). They may also impact standard cognitive processes; eg, memory and attention are likely to be disrupted when thoughts seem alien or foreign, and are no longer felt to be a relatively transparent means of apprehending the world.18

Some of these cognitive phenomena bear resemblance to the concept of “soundless voices,” which can be traced back to Bleuler’s observations of patients with hallucinations: they did not describe the sensory features of actually “hearing” voices but related them to cognitive intrusions. Those who experience these phenomena are often convinced of the “otherness” of such intrusions and tend not to attribute the origin of these soundless voices to their own thinking processes.19 On the other hand, thought echo or “audible thoughts” denotes the perceptualization or vocalization of one’s own thoughts into an external space, whereas thought insertion is the inward projection of “other” thoughts into one’s mental space. All three phenomena can be framed as alterations of thought-agency and ownership.20

Along similar lines, Mayer-Gross proposed that thought insertion involves a “becoming sensory” (Versinnlichung) of those thoughts.21 That is, they are no longer experienced as thoughts but as material objects inserted by foreign agency. He further noted that in early schizophrenia there may be the nearly simultaneous experience of verbal and nonverbal hallucinations (akoasms such as buzzing, whistling, and roaring), thought insertion, thoughts becoming loud, etc. Thus, it has been argued that auditory verbal hallucinations and disturbances of thought agency and ownership differ in degree rather than in kind.22 Such theories suggest that focusing on the easily measurable “auditory verbal hallucinations” may obfuscate these subtler variants; more research is necessary to determine whether a broader definition is necessary to encompass these and other variations of hallucinations,14 and whether they should be viewed as on a continuum with nonpsychotic phenomena.

Perceptual Experience

Although the above section discusses the semi-perceptual qualities of cognition, an additional line of research focuses on perceptual alterations themselves. Heidelberg psychiatrist Mayer-Gross may be regarded as herald of the “perceptual anomalies” approach to schizophrenia. Mayer-Gross, his Heidelberg contemporary, Viktor von Weizsäcker, and the phenomenological psychiatrists Matussek, Conrad, Binswanger, and Blankenburg all considered changes in perceptual organization to play a fundamental role in schizophrenia.21,23 Matussek, eg, emphasized the impact of disruptions of context on the perception of objects, which allows certain perceptual qualities/details to dominate over others and be imbued with unusual significance.24 Rather than circumscribed perceptual alteration, Matussek highlighted a global, Gestalt-like change in the structure of experience.

Mayer Gross’ fellow colleague in Heidelberg, Jaspers25 developed the concept of “pseudohallucinations” (PH), which stood in stark contrast to the perceptual anomalies approach.26 Jaspers adapted his concept from the German psychiatrist Hagen and the Russian psychiatrist Kandinsky (who himself experienced hallucinations). Not considered “real” hallucinations, PH are not perceptual experiences but based on ideas (Vorstellungen). For Jaspers, genuine or true hallucinations are experienced as “objectively” present (leibhaftig), and may be experienced alongside real perceptions in external objective space. In contrast, PH are imaginary (bildhaftig), not experienced as concretely real, have the character of subjectivity, and appear in inner subjective space. PH have been criticized for conceptual confusion, lack of clinical utility, and historical inaccuracy.26–28 Mayer-Gross criticized Jaspers’ overly strict opposition between “objective” perception (genuine hallucination) vs subjective image (PH), or “outer” vs “inner,” which often failed to distinguish actual clinical cases.26 It also does not leave open the possibility that PH may also have their source in low level sensory or sensorimotor processing.

Mayer Gross’ approach is supported by a number of empirical studies, which have found evidence of fragmentation, or changes in perceptual organization, in schizophrenia, including difficulties seeing objects or scenes as whole gestalts.29–31 Perceptual fragmentation is associated with worse outcomes,31 and may contribute to figure-ground confusion and loss of perceptual stability, with objects appearing to change shape or appearance.32,33 Persons with these visual distortions are also more likely to experience visual hallucinations,34 possibly pointing to shared processes contributing to these phenomena. These lines of research suggest that sensory and perceptual anomalies, including perceptual disorganization, may be implicated in phenomena that meet criteria for hallucinations, and investigation of these changes and the processes involved may shed light on the development of hallucinations.

Selfhood and Reality

Much typical human experience consists of being engaged and absorbed in a combination of cognitive, physical, and affective activity within a world of (animate and inanimate) objects. The first-person perspective or “mineness” of this interaction with the world provides a form of implicit self-awareness (it is I who is having these experiences), which is sometimes referred to as self-presence or as self-affection.35 This can be altered in the schizophrenia spectrum, in which the self may seem to stand distant or alienated from experience, rather than implicitly and constantly present.18 Various forms of depersonalization, a sense of inner void, and derealization, a feeling that objects or the world are less real or immediately present, can reflect the weakening of self-presence.16 Numerous studies have found that changes in self- and reality-experience in the form of dissociation predict positive psychotic symptoms, including hallucinations, particularly among individuals who have experienced childhood trauma.36

Explanations for the relationship between dissociation and hallucinatory phenomena are varied, and include the hypotheses that perceptual anomalies increase the risk of dissociation37; that dissociation weakens cognitive inhibition and therefore increases vulnerability to anomalous perceptual experiences36; and that hallucinations may be features of dissociation (ie, inner speech is experienced as separate from oneself).38 Such findings may challenge the notion of the schizophrenia spectrum and even psychosis itself (ie, that such symptoms do not reflect a disturbance of reality testing, but are better understood as dissociative processes),39 although other factors unique to schizophrenia or psychosis may also play a role in increasing vulnerability to hallucinations. Alterations of selfhood and presence, then, may be contributing factors in the development of hallucinations across multiple distinct psychiatric conditions.


Persons with psychosis may experience a variety of time distortions, such as lack of sense of time continuity, or a dissociation between internal and external time, ie, a mismatch between the speed of inner awareness and the speed of external events (supplementary material note iv).40–47

Some phenomenologists have theorized that alterations in the sense of time continuity is associated with changes in “minimal” self, ie, the ability to feel oneself as being present here and now.43,46 Giersch and Mishara, eg, have proposed that disruption to minimal self in schizophrenia involves abnormalities in unconscious automatic processing, including the processing of time (at intervals far too brief to be experienced consciously; supplementary material note v).48–50 

Experimental research indicates that anomalies of temporal processing in schizophrenia patients emerge at very brief time scales (8 and 17 ms),51 suggesting that “protentions” (ie, anticipations [often nonconscious] that serve as the context for moment-to-moment predictions) are altered in psychosis and occur very early in the “phenomenological hierarchy” (supplementary material note vi).

These results have been proposed to be consistent with impairments in predictive coding,52 the notion that sequences of information are predicted in advance to allow for the detection of unexpected events and a fluid perception of the outer world.51 In addition, Fuchs has argued that the alterations in selfhood in schizophrenia are related to changes in the ability to link the present moment with what is about to happen, and to move into the probable or anticipated future, which may result in the disconnection and fragmentation of perceived events.53

This work on continuity of minimal self points to the possible role of temporal changes in alterations in cognition and perception, such as the sense of ownership and anticipation of self-generated events, thought to be related to hallucinations: when thoughts and sensations are no longer anticipated and embedded in a sense of self that persists over time, they may instead be experienced as externally generated objects rather than “inhabited” aspects of selfhood.

Interpersonal Experience

Alterations in social experience have long been one of the hallmarks of schizophrenia and play a key role in functional outcome. Many phenomenological psychiatrists have emphasized changes in interpersonal attunement and apparent alterations in the commonsense social world in schizophrenia.

Bleuler considered autism, the “detachment from reality, together with the relative and absolute predominance of the inner life,”54 to be highly specific to schizophrenia, while Minkowski proposed that a “loss of vital contact” with the immediate, social world was the core alteration underlying schizophrenia.55 Indeed, many persons with first-episode schizophrenia describe the sense of being different from others,56 which cannot be articulated in terms of concrete characteristics, but rather as having more of an ontological dimension57; as one individual described it, “I always felt different, as if I belonged to another race.”58

 In addition, psychosis has long been viewed as an alteration in the boundary between self and other. Laing suggested that persons with psychosis are prone to feelings of “engulfment,” the threat of losing one’s identity to the influence of others, and believed that the core alteration in schizophrenia could be characterized as “ontological insecurity,” or a fundamental difficulty regarding differentiation with the world, autonomy, and continuity of being.59

Alterations of both attunement and ego-boundaries may be involved in hallucinatory experiences. Federn,60 eg, suggested that psychotic experiences including hallucinations and delusions were indicative of changes in the inner ego boundary, the dynamic interplay between “inner mentality” and external reality or “non-ego.”

In addition, some individuals with psychosis describe a tendency to prioritize private (not intersubjectively available) perceptual phenomena and to trust the reality of those phenomena over consensual, intersubjective reality, while others have described a confusion between what is “real” (ie, perceived by others) and what is imagined, remembered, or dreamed.61 

Research on source-monitoring in schizophrenia—the ability to determine or remember whether a stimulus was internally or externally generated—lends some support to this hypothesis.62,63 

Furthermore, research on the impact of interpersonal experience on hallucinations suggests that interpersonal context may be at least as important as—and potentially contribute to—perceptual or cognitive transformations; and that therapeutic work on relationships and communication with others and with voices may help individuals cope with hallucinations.64,65


Alterations in bodily self-representation have been suggested to be another core component of schizophrenia, and include sudden changes in size and shape of the body, alterations in body ownership, anomalous agency, and even out-of-body experiences.33,66,67 Cenesthesic changes (involving unusual awareness of or changes in bodily sensations), including numbness of the body, electric and thermal sensations, sensation of abnormal pressure or weight, and vestibular sensations, are also observed in schizophrenia and are associated with clinical symptoms.68 Such changes in bodily self-presence are present during the prodromal stage and remain salient throughout the course of schizophrenia.18,69–73

Although these phenomena may contribute to a range of hallucinatory somatic experiences, some of the most striking phenomena appear as forms of “autoscopy.” Autoscopy is a loosely related complex of experiences in which one sees (or experiences) a “double” as external to one’s current vantage point.

The autoscopic experience may last from seconds to hours or, in some cases, be present for years at a time (supplementary material note vii).74 In vulnerable individuals, autoscopic experience can be triggered via a proprioceptive-tactile illusion,75 while in the general population, autoscopic phenomena are associated with elevated schizotypy75,76 and transliminality (a hypothesized tendency toward multisensory experiences, mystical experience, and absorption).77,78 Some voice hearers have indicated that their voices have much more of a physical, embodied quality, rather than being purely auditory (or of another sensory modality).79 

Such findings on alterations of embodiment and autoscopic experiences suggest the need to focus on the relationship of changes in mental representations of the body to the development of a range of hallucinatory experiences.


We have considered a range of experiential features and hypothesized mechanisms associated with psychosis, with attention to one commonly researched feature of psychosis, hallucinations. In doing so, we hope to have demonstrated not only the variety of manifestations of psychosis, but also the intricate intertwining of experiential domains and the interplay between phenomenology, neurocognitive factors, and environment.

It is unlikely that hallucinations or any other experiential alterations traditionally associated with psychosis will be experienced as discrete or static phenomena. Indeed, the findings presented above suggest that hallucinations may occur in multiple perceptual modalities and may be continuous with nonhallucinatory experiences (eg, thought insertion, out-of-body experiences, dissociation, alterations in perception).

It is similarly unlikely that the development of hallucinations can be attributed to one core process or causal factor. Genetics, neurocognitive processes, subjective experience, cognitive styles or patterns of interpretation, and cultural and social environments are likely to interact in complex ways, with hallucinations as an equifinal outcome.

Given this complexity, it is understandable that research, diagnostic criteria, and clinical interventions would attempt to simplify the defining characteristics of hallucinations. However, doing so risks ignoring a diverse array of associated or interwoven experiences, focusing instead on symptoms with narrow, concrete, and easily agreed-upon definitions.

This review suggests some key ways in which research on hallucinations can evolve. First, there are a number of identifiable subjective experiences co-occurring with hallucinations that do not fit the conventional definition of hallucinations.

These are in need of greater research attention, both on their own and in relation to hallucinatory phenomena. As part of this working group, ongoing and future projects will systematically inquire into the full range of experiential modalities that are involved in participants’ hallucinations, while other investigations may be facilitated by wide-ranging phenomenologically oriented interviews.16,47

 Second, patterns of covariation with hallucinations need to be understood, and if clustering together, explanatory and etiological models of hallucinations need to be able to account for their co-occurrence. For example, several researchers are already exploring the relationship between phenomena like thought insertion and voice hearing.20,115 In addition, dissociation among voice hearers may moderate various outcomes associated with hallucinations; addressing this factor may reduce unnecessary variability within samples.

Third, the breadth of phenomena suggests reconsidering how hallucinations should be defined. Traditional definitions have emphasized sensory qualities and their realness and distinctiveness from mental imagery and verbal thought. The importance of this boundary is questionable when considering findings on this range of experiential transformations. For example, future research on hallucinations may also benefit from including subjects who experience thought insertion or thought passivity (rather than limiting recruitment to those who meet a narrower definition of hallucinations). There may be value in incorporating other elements into a definition of a broader construct of hallucinations, such as changes in agency or ownership of thought, separation from self-experience, uncertainty about internal vs external phenomena, and anomalous awareness of the presence of others.

Clinical Implications

A potential benefit of broadening the lens of hallucinations is to assist practitioners in attending to, and developing a vocabulary for enquiring about, these broader aspects of experience. Interventions targeting auditory hallucinations have focused on typical verbal characteristics, and the common perception of voices as sentient others.116 These have included cognitive restructuring targeting beliefs about voice power,117 and reducing submissive and hostile interpersonal relationships that develop with voices.64,118 While common, these targets are rather specific, and potentially secondary, elements of hallucinations rather than the more fundamental phenomenology of psychosis.

In considering this broader phenomenology, therapeutic approaches may try to more directly target the underlying changes in areas such as self experience and sensory and somatic integration. To date, there has been some development of methods to address depersonalization and derealization experiences in people with psychosis,119 but these remain at an early stage. Approaches that encourage metacognitive awareness may also be helpful. For example, Lysaker and colleagues120 have developed interventions designed to foster awareness of different cognitive operations and integrate them into a more coherent experience of self. Finally, this review suggests the need for further investigation of therapies that target the interactive, dynamic, and phenomenologically rich nature of hallucinations, such as Avatar Therapy,121 Compassion Focused Therapy,122 and Voice-Dialogue.123


This article has reviewed domains of experience that extend beyond voices to include self and reality, cognitive experience, perceptual anomalies, temporality, interpersonal experience, and embodiment. In addition, we have reviewed several of the major theories of prereflective change in psychosis to describe the ways these experiential transformations are inter-related.

We also considered the impact of trauma on the development of hallucinations, though additional work is necessary to more fully consider the impact of cultural, social, and environmental factors on the construct of psychosis and the expression of associated symptoms.

We suggest that all of these domains should be considered in directing future research or clinical work on hallucinations—or any other aspect of psychosis. While symptom-specific research has resulted in valuable new discoveries and clinical interventions, it is essential not to lose sight of the experiential context out of which hallucinations develop.

Mount Sinai Hospital
Media Contacts:
Elizabeth Dowling – Mount Sinai Hospital
Image Source:
The image is in the public domain.

Original Research: Open access
“Abnormal auditory tonotopy in patients with schizophrenia”. Gaelle E. Doucet, Maxwell J. Luber, Priti Balchandani, Iris E. Sommer & Sophia Frangou.
NPJ Schizophrenia doi:10.1038/s41537-019-0084-x.


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.