Research finds stronger brain responses to inner speech in individuals with auditory verbal hallucinations — Evidence Review
Published in Schizophrenia Bulletin, by researchers from UNSW Sydney
Researched by
— the AI search engine for science
Table of Contents
A new study finds that in schizophrenia, the brain may misidentify a person's own inner voice as coming from the outside world, supporting long-standing theories about the origins of auditory hallucinations. Most related research agrees with these findings, showing consistent brain network disruptions in those who experience auditory verbal hallucinations (UNSW Sydney).
- Multiple neuroimaging studies have identified abnormal activity in speech and auditory processing regions, such as Broca's and Wernicke's areas, in people experiencing auditory hallucinations, supporting the idea that self-generated thoughts may be misinterpreted as external voices 1 3 6.
- Research into the neural mechanisms of auditory verbal hallucinations (AVH) consistently points to failures in self-monitoring and reality-testing networks, with some studies also implicating disrupted connectivity between key brain regions involved in speech evaluation 3 6 14.
- There is general consensus that while the phenomenology of inner speech may appear similar between patients and controls, functional brain responses and connectivity patterns during AVH are distinct in clinical populations, with some evidence of a continuum in non-clinical voice-hearers 4 7 11 13.
Study Overview and Key Findings
Auditory verbal hallucinations—or "hearing voices"—are among the most prominent and distressing symptoms of schizophrenia, yet the biological mechanisms underlying them remain unclear. The new research led by Professor Thomas Whitford at UNSW Sydney addresses a critical gap in the field by directly measuring brain responses to inner speech in both healthy individuals and people with schizophrenia, using a non-invasive EEG approach. This work is significant not only for advancing understanding of AVH, but also for identifying potential biomarkers that could improve early detection and intervention for psychosis, which currently lacks objective diagnostic tools.
| Property | Value |
|---|---|
| Organization | UNSW Sydney |
| Journal Name | Schizophrenia Bulletin |
| Authors | Professor Thomas Whitford |
| Population | People with schizophrenia spectrum disorders |
| Sample Size | n=142 (55 with AVH, 44 without AVH, 43 healthy) |
| Outcome | Brain responses to inner speech and auditory hallucinations |
| Results | People with AVH showed stronger brain reactions to inner speech. |
Literature Review: Related Studies
To situate these findings in the broader research context, we searched the Consensus paper database (over 200 million papers) using focused queries. The following search queries were used:
- auditory verbal hallucinations brain activity
- inner speech perception AVH correlation
- neurobiology of voice hearing experiences
Below, we summarize key topics and findings from the most relevant studies:
| Topic | Key Findings |
|---|---|
| What brain regions and networks are implicated in auditory verbal hallucinations? | - AVH are consistently associated with increased activation in fronto-temporal areas, including Broca's area, insula, and superior temporal gyrus, as well as the medial temporal lobe 1 3. - Abnormal activation and connectivity in the salience and default mode networks are linked to AVH, with impaired integration among auditory, language, and memory regions 5. |
| How does inner speech relate to the experience of hearing voices? | - Failures in self-monitoring of inner speech may cause self-generated thoughts to be misattributed as external voices, with impaired connectivity between speech evaluation regions such as the superior temporal and anterior cingulate cortex 6 14. - Phenomenological reports of inner speech are similar between patients with AVH and controls, but neural responses differ 7 9. |
| Are hallucinations in psychosis and non-psychotic individuals similar or distinct? | - Brain activation during AVH is highly similar in psychotic and nonpsychotic individuals, suggesting a shared cortical network, though clinical voice-hearers experience more distress and negative content 4 11 12. - Non-clinical voice-hearers may exhibit enhanced attention and perceptual processing to ambiguous speech, implicating a continuum rather than a dichotomy 11 13. |
| What cognitive and perceptual processes underlie hallucinations? | - Top-down perceptual failures and overweighting of perceptual priors can contribute to hallucinations, with cognitive biases and impaired reality monitoring implicated in both clinical and non-clinical voice hearers 10 13 14. - Subjective qualities like loudness and reality of AVH may relate to the degree of activation in inner speech and right hemisphere language areas 9. |
What brain regions and networks are implicated in auditory verbal hallucinations?
A robust body of neuroimaging research demonstrates that auditory verbal hallucinations activate a distributed network of brain regions involved in speech generation, perception, and memory. This aligns with the new study's findings of abnormal responses in the auditory cortex during inner speech tasks in schizophrenia. The dysfunction appears to extend beyond isolated regions, involving broader network-level disruptions, particularly in fronto-temporal and salience networks.
- Meta-analyses have repeatedly identified increased activity in Broca's area, insula, superior temporal gyri, and medial temporal lobe during AVH episodes 1 3.
- Abnormal functional connectivity between salience and default mode networks is observed in first-episode psychosis patients with AVH 5.
- Activation patterns during AVH often differ in lateralization from those seen in normal inner speech, with right inferior frontal areas more engaged during hallucinations 2.
- Memory-related regions, such as the medial temporal lobe, are also consistently implicated, suggesting a possible role for memory intrusions in AVH content 1 3.
How does inner speech relate to the experience of hearing voices?
The theory that AVH result from a breakdown in self-monitoring of inner speech is strongly supported by both experimental and neuroimaging evidence. The new study's EEG findings directly demonstrate altered brain responses to inner speech in people experiencing hallucinations, providing one of the most direct tests of this theory to date.
- Patients with AVH show impaired functional integration between the superior temporal and anterior cingulate cortex when distinguishing their own speech from that of others 6.
- Failures of top-down cognitive control over perceptual processes can result in self-generated speech being misattributed as external 14.
- Despite nearly identical subjective reports of inner speech between patients with AVH and healthy controls, functional brain responses to inner speech are disrupted in clinical groups 7 9.
- The degree of neural activation in inner speech processing networks may relate to the subjective loudness or "reality" of the hallucinations 9.
Are hallucinations in psychosis and non-psychotic individuals similar or distinct?
There is significant evidence that AVH are not exclusive to psychosis, with non-clinical voice-hearers exhibiting brain activation patterns similar to those with schizophrenia during hallucinations. However, differences in phenomenology, distress, and the ability to control hallucinations suggest important distinctions, supporting a continuum model.
- Both psychotic and nonpsychotic individuals with AVH activate similar cortical networks during hallucinations, with no significant differences found in key regions 4.
- Non-clinical voice-hearers, such as clairaudient psychics, report less distress and more control over their experiences compared to clinical populations, despite similar neural activation 12.
- Healthy voice-hearers show enhanced neural responses to ambiguous speech, possibly reflecting greater reliance on prior knowledge and expectation in perception 11 13.
- Differences in beliefs, emotional content, and distress appear to differentiate clinical from non-clinical AVH, even when underlying brain activity is similar 11 12.
What cognitive and perceptual processes underlie hallucinations?
Research increasingly highlights the role of cognitive biases, top-down processing failures, and the overweighting of prior expectations in the genesis of hallucinations. The new study builds on this by showing that disrupted predictive coding in the auditory cortex may underlie the misattribution of self-generated speech.
- Conditioned hallucinations can be induced via Pavlovian learning, with voice-hearers more susceptible to these perceptual errors, implicating cognitive processes in hallucination risk 10.
- Impaired reality monitoring and reduced top-down control from prefrontal areas may prevent individuals from correctly attributing inner speech to themselves 13 14.
- The complexity and subjective qualities of AVH, such as loudness or external location, may reflect the degree of engagement in right hemisphere language areas and variability in inner speech network activation 9.
- Both clinical and non-clinical voice hearers may benefit from a unified theoretical framework that accounts for perceptual, cognitive, and neurobiological processes 3 13.
Future Research Questions
While recent studies have advanced understanding of the neural mechanisms underlying auditory hallucinations, several important questions remain. Future research is needed to clarify the specificity of these findings, their relevance across the psychosis spectrum, and their potential for early detection, intervention, and treatment development.
| Research Question | Relevance |
|---|---|
| Can EEG-based biomarkers of inner speech prediction errors predict psychosis onset? | Identifying reliable EEG markers could enable earlier detection and intervention for individuals at risk of psychosis, addressing a major gap in current clinical practice 5. |
| How do inner speech mechanisms differ between clinical and non-clinical voice-hearers? | Understanding similarities and differences in brain network function and phenomenology could clarify why some individuals develop distressing AVH while others do not 4 11 12. |
| What is the role of memory intrusions in the content of auditory verbal hallucinations? | Medial temporal lobe involvement suggests that AVH content may in part derive from intrusive memories, an area warranting further investigation to inform therapies 1 3. |
| Can targeted interventions restore normal predictive coding in the auditory cortex to reduce AVH? | If prediction errors underlie AVH, cognitive or neuromodulation interventions aimed at restoring predictive processing could provide novel treatments 10 14. |
| How do emotional and semantic characteristics of voices relate to neural activation patterns during AVH? | Exploring the neural correlates of AVH phenomenology (e.g., emotional valence, semantic complexity) could inform personalized interventions and deepen understanding of symptom variability 2 9 12. |