Behind Hallucinations – reasons and mechanisms

BotX team
October 16, 2023

Caught up in the fascination of observing hallucinations, we may forget, for a second, that they are no more than results of some intricate processes taking place in the brain. There must be a reason and a way in which they appear, an underlying mechanism, even if we haven’t quite figured it out yet.

Let’s start from the beginning. In a previous article we’ve said that, by definition, having a hallucination is perceiving something that is not there. But how does the mind do that?

We may expect the processing of a real perception to be massively different from that of a hallucination, or maybe even for the brain itself to be differently structured or wired.

Different brain structure?

An fMRI of a brain that is having an auditory hallucination is not that different from one that hears an actual voice – with significant involvement of the primary auditory cortex and language area. 1 Similarly, a visual hallucination will trigger the occipital and temporal cortices. Such observations led many to believe that a hallucination is a self-generating, distorted, perception; one that alters the normal course of sensory processing, but takes place in a similar place and manner as a perception nonetheless.2

On the other hand, other regions such as the hippocampus and the parahippocampal gyrus also exhibit high levels of activity during auditory or visual hallucinations – suggesting memory plays a significant role in the process. The insula, a region important for self-awareness and perception of others, is also active during a hallucination, particularly in schizophrenia patients. 2

Furthermore, studies taking a closer look at the actual structure of different brain regions found significant differences. Namely, differences in gray matter volume, not just between people with and without hallucinations, but also between patients with different psychiatric diseases. One study shows that schizophrenic patients tend to have decreased gray-matter volume in areas such as the precentral gyrus, caudate body, and cerebellum. In contrast, in bipolar patients, different regions are affected, such as the middle frontal gyrus, fusiform gyrus, and post-central gyrus. 3 Another one correlated how prone schizophrenic patients were for hallucinations with the length of the paracingulate sulcus, a fold towards the front of the brain.4 Yet another found a connection between hallucination proneness and the microstructure of the superior temporal gyrus, specifically the complexity of the neurons’ dendrites in the area.5


The main treatment for hallucinations is antipsychotics – medications that decrease the amount of dopamine available at synapses. Initially discovered by serendipity, they brought about the dopamine hypothesis of psychosis – the idea that hallucinations are triggered by excessive amounts of dopamine in the brain, specifically in the striatum. Although it is a long-standing view in psychiatry, it proves, at best, incomplete.6

A severe dopamine deficit, for example, can also trigger hallucinations. Patients with Parkinson’s disease (in which dopamine-producing neurons are progressively destroyed) may experience hallucinations even before they are treated with dopamine-like substances. As is often the case, equilibrium is necessary for normal function. 7

Furthermore, other neurotransmitter imbalances have been associated with hallucinatory events. Acetylcholine, for example: It is long known that plants such as belladonna (that inhibits acetylcholine transmission) can cause alterations of consciousness and hallucinations. Serotonin plays a role as well – with hallucinogenic substances such as LSD, mescaline, psilocybin and ecstasy acting largely on serotonin receptors. The same can be said for glutamate – with glutamate inhibitors, such as ketamine and phencyclidine, also inducing hallucinations. Ultimately, diminished GABA transmission was also correlated with the intensity of hallucinations.7

It is becoming clear that dopamine cannot be the sole unifying theory of hallucinations. Instead, we must take into account various factors including genetic predispositions, multiple neurochemicals, diverse structural alterations, as well as psychological and social aspects when tackling such a complicated situation. 6

Results of an over-expectant brain?

“We predict ourselves into existence”, argues neuroscientist Anil Seth in his fascinating TED talk, ‘Your brain hallucinates your conscious reality’.

We’ve seen before how some brain processes, memory being the prime example, involve a great deal of creation and prediction rather than exact reproduction of the original information. Stored information is adapted and modulated to produce an echo of the original story, one filtered through current perspectives and beliefs.

The same can be said, to some extent, about perception. We constantly perceive the external world with some degree of prediction. Otherwise, it’d be impossible for us to catch a ball that’s being thrown at us. We also perceive with some slight level of expectation, based on all our accumulated knowledge and previous similar experiences. We know that snow is cold before we touch it, just as well as we know the texture of sand before we step on it.

One study argues that this is where hallucinations stem from: An over-expectant brain. One that is so sure an actual sensation (a sound, for example) is about to be heard, that it actually hears it. Researchers induced a ‘ conditioned hallucination’: employing Pavlovian mechanisms, they showed subjects a checkboard image and simultaneously played a sound. In a second stage, the subjects were presented with the image alone. Conditioned hallucinations occurred when subjects still reported hearing the sound upon seeing the checkboard pattern.

The study compared psychotic patients (that heard voices or not) with psychics (people who heard daily voices but did not receive diagnosis or treatment, as they attributed them metaphysically) and controls who did not hear voices.

We’ve seen that hallucinations in otherwise healthy individuals are not an uncommon feat. As such, it’s not surprising to find that all groups experienced conditioned hallucinations. However, people who reported daily hallucinations (regardless of whether they were diagnosed as psychotic or not) reported more conditional hallucinations, and were significantly more confident in their answers.

Imaging studies performed on the same subjects showed activity in the same areas that are usually involved in auditory hallucinations. However, the fact that they lighted up just slightly before actually being stimulated makes them resemble a prediction rather than a detection.

Ultimately, what the study aims to suggest is that there may be some form of a continuum from mental health to illness, one that relies on prediction. 8,9

Memory intrusion?

We’ve seen that areas such as the hippocampus and parahippocampal gyrus seem to light up during hallucinations consistently. This may suggest that a hallucination is, in fact, a failure to suppress an intrusive thought or memory.

Other results supporting this include the fact that both patients with schizophrenia and healthy individuals that report hallucinations seem to do worse on memory tests. Perhaps a hallucination is, in essence, a memory that is not stored correctly. Taken out of context, maybe arising spontaneously, it may appear foreign, or even go as far as to enter consciousness in an intrusive manner. 2

Estranged inner monologue?

When we speak with “the voice in our heads” we seldom refer to an actual autonomous voice. It’s rather the flow of conscious thoughts that accompany our every action. For some, this inner narrative is more substantial, more vibrant, and may even associate the idea of an actual voice speaking.

Some suggest auditory hallucinations may simply be an inner monologue that has gone estranged, and seems to be coming from the outside. This is consistent with the fact that voices heard in psychosis refer the person directly, often with an accusing or criticizing tone.

One study even reported that as much as 78% of voice-hearers could identify whom the voice belongs to – often a family member or past abuser. 10

In another study, 35 healthy individuals were asked to determine whether a set of pre-recorded messages were voiced by them or not. Results showed that subjects with a higher predisposition for hallucinations had a harder time recognizing their own voice or telling apart their voice from another’s, regardless of the content of the message or its emotional valence. 11

According to this view, the actual deficit resides in reality-monitoring or self-monitoring systems, which in turn make internal thoughts seem foreign. What prompted this hypothesis was the abnormal activity of the insula, observed in hallucinating subjects. Given the proposed role of the insula in self-representation and perception of other ‘selves’, it may be that hallucinations, auditory ones in particular, reflect an inability to recognize that a specific experience has an internal rather than external source. 2

Another region of interest is the medial prefrontal cortex observed during hallucinations –involved in reality monitoring ability. Studies have described low levels of activity in this region in individuals prone to psychosis, as well as in schizophrenia patients. However, it remains unclear whether this could cause the hallucinations, or just accompanies them. 2

As evidence for this theory accumulated, it got some to believe the answer to hallucinations may then reside in correctly stimulating the reality-monitoring areas in the brain, giving rise to attempts at neurostimulation as a treatment option for auditory hallucinations. Although it seemed like a promising alternative, results so far were modest, and much work needs to be done in the area before a conclusion can be reached. 12

Ultimately, hallucinations remain an elusive phenomenon. We can only hope that joint efforts from various fields, including but not limited to psychiatry, neuroscience, psychology, and even digital technology 13 will aid us in unraveling their intricacies.


  1. Thoma, R. J. et al. Functional MRI Evaluation of Multiple Neural Networks Underlying Auditory Verbal Hallucinations in Schizophrenia Spectrum Disorders. Front. Psychiatry 7, 39 (2016).
  2. Zmigrod, L., Garrison, J. R., Carr, J. & Simons, J. S. The neural mechanisms of hallucinations: A quantitative meta-analysis of neuroimaging studies. Neuroscience and Biobehavioral Reviews 69, 113–123 (2016).
  3. Not All Hallucinations Are Alike: The Brain Tells Us So | Psychiatric Times. Available at: link. (Accessed: 4th June 2020)
  4. MRI scans show hallucinations linked to differences in brain structure. Available at: link. (Accessed: 4th June 2020)
  5. Spray, A., Beer, A. L., Bentall, R. P., Sluming, V. & Meyer, G. Microstructure of the superior temporal gyrus and hallucination proneness - a multi-compartment diffusion imaging study. NeuroImage Clin. 20, 1–6 (2018).
  6. Teufel, C. Sensory Neuroscience: Linking Dopamine, Expectation, and Hallucinations. Current Biology 28, R158–R160 (2018).
  7. Chaudhury, S., Kumar, S. & Soren, S. Hallucinations: Etiology and clinical implications. Ind. Psychiatry J. 18, 119 (2009).
  8. Powers, A. R., Mathys, C. & Corlett, P. R. Pavlovian conditioning–induced hallucinations result from overweighting of perceptual priors. Science (80-. ). 357, 596–600 (2017).
  9. Do You Hear What I Hear? Auditory Hallucinations Yield Clues to Perception - Scientific American. Available at: link. (Accessed: 1st June 2020)
  10. Corstens, D. & Longden, E. The origins of voices: Links between life history and voice hearing in a survey of 100 cases. Psychosis 5, 270–285 (2013).
  11. Pinheiro, A. P., Farinha-Fernandes, A., Roberto, M. S. & Kotz, S. A. Self-voice perception and its relationship with hallucination predisposition. Cogn. Neuropsychiatry 24, 237–255 (2019).
  12. Moseley, P., Fernyhough, C. & Ellison, A. Auditory verbal hallucinations as atypical inner speech monitoring, and the potential of neurostimulation as a treatment option. Neuroscience and Biobehavioral Reviews 37, 2794–2805 (2013).
  13. Potential Applications of Digital Technology in Assessment, Treatment, and Self-help for Hallucinations | Schizophrenia Bulletin | Oxford Academic. Available at: link. (Accessed: 4th June 2020)

BotX team