Read this topic

When a patient describes hearing, seeing, or feeling something that is not really there, the instinct of a fresh MBBS graduate is to reach for "psychosis" immediately. That instinct is often wrong. The exact subtype of a hallucination matters because the sensory modality involved, and its precise quality, points toward very different likely causes, from a primary psychotic illness to a neurological problem to simple substance withdrawal. For example, imagine two patients seen back to back in clinic: one hears two voices discussing his actions in the third person, the other sees small animals crawling across her bedroom floor. Both are "hallucinating," but getting the pattern right changes the whole diagnostic pathway that follows.

A hallucination (a false perception experienced with no real external stimulus present, but felt with the same force and conviction as a genuine perception, and not under voluntary control) is different from an illusion (a misperception of a real stimulus that IS actually present, mistaking a coat on a hook for a person in a dark room). Hallucinations can occur in any sensory modality: auditory (hearing), visual (seeing), olfactory (smell), gustatory (taste), and tactile (touch). Each modality carries a different differential diagnosis, which is why naming the modality precisely is often the single most useful first step in a clinical assessment, far more useful than simply writing "hallucinations present" in the notes.

The trap most beginners fall into is treating every unusual perceptual experience as the same thing. Three experiences get lumped together that shouldn't be: an illusion (a real stimulus, misperceived), a true hallucination (no real stimulus, experienced as coming from the external world, with no insight that it is unreal), and a pseudohallucination (perceived in inner, subjective space, inside the mind's eye or ear, with the patient retaining insight that the experience is not real). Insight is the discriminator that trips people up most: a person with a pseudohallucination knows it isn't real even while experiencing it vividly, while a person with a true hallucination does not. This single distinction is worth more marks in an exam than any amount of memorised modality lists.

The reason modality is so diagnostically powerful comes down to which part of the brain is misfiring. Auditory processing, visual processing, and the circuits involved in insight and reality-testing are anatomically and functionally distinct, so a lesion, drug, or disease process that disrupts one pathway tends to produce hallucinations in that pathway's modality rather than randomly across all senses. A psychotic illness like schizophrenia disrupts the brain's broader belief- and salience-forming networks, which is why it most often produces auditory hallucinations with rich, personally meaningful content (voices commenting or discussing). An organic insult (a seizure focus, a toxin, visual pathway damage) tends to produce hallucinations tied to whichever sensory circuit sits nearest the pathology. Once you accept that logic, the rest of this topic becomes pattern-matching rather than memorising a list.

Auditory hallucinations predominate in schizophrenia, especially voices commenting on the patient's actions or discussing the patient in the third person. Visual hallucinations more often point toward an organic cause: delirium (an acute confusional state, usually from an infection, drug, or metabolic disturbance), dementia with Lewy bodies, or Charles Bonnet syndrome (vivid visual hallucinations in visually impaired patients who retain full insight and have no psychiatric illness at all). Olfactory and gustatory hallucinations, such as smelling burning toast with no source, classically point toward temporal lobe epilepsy. Tactile hallucinations of insects crawling on the skin, sometimes called formication, point toward stimulant intoxication or alcohol withdrawal. Two rarer patterns are worth knowing precisely: a functional hallucination occurs when a real external stimulus and the hallucination occur in the same modality at the same time (for example, hearing voices only while a tap is running), while a reflex hallucination occurs when a stimulus in one modality triggers a hallucination in a different modality (a sound triggering a visual hallucination). An extracampine hallucination is perceived as occurring entirely outside the normal range of the relevant sensory field, such as "hearing" a voice from another city.

The exam-favourite confusions cluster around three pairs. First, illusion versus hallucination: an illusion always has a real stimulus behind it (mistaking a curtain for a figure), while a hallucination has none. Second, true hallucination versus pseudohallucination: identical sensory content can be reported by two patients, but only their insight and the perceived location (external world versus inside the mind) tell you which one it is. Third, functional versus reflex hallucination: both involve a real stimulus in one modality, but a functional hallucination stays in the same modality as the trigger, while a reflex hallucination crosses into a different modality. Two more normal-variant look-alikes deserve a mention so they are never mistaken for illness: hypnagogic hallucinations occur at the transition into sleep, and hypnopompic hallucinations occur at the transition out of sleep on waking; both are usually normal phenomena in an otherwise well person and are not, by themselves, evidence of a psychotic illness.

There is no blood test for "hallucination type": it is established through a careful phenomenological history, not a scan. The clinician needs to ask, precisely: which sense is involved, does the experience come from outside or inside the head, does the patient believe it is real (insight), is there any trigger stimulus present, and what is the content and context (mood, sleep state, substance use, physical illness)? A collateral history from family is invaluable, especially for the fluctuating alertness seen in delirium or dementia with Lewy bodies. Once modality and context suggest an organic cause, targeted investigation follows: bloods (infection markers, electrolytes, glucose, thyroid function, B12), a physical examination looking for signs of intoxication or withdrawal, and, where epilepsy or an acute neurological cause is suspected, imaging or an EEG (electroencephalogram, a recording of the brain's electrical activity) to look for a temporal lobe focus. Cognitive screening and a careful sleep and substance history complete the picture.

Management depends entirely on the underlying diagnosis, so accurate discrimination is not academic; it directly shapes treatment. Where hallucinations arise from a primary psychotic illness such as schizophrenia, first-line management follows NICE guidance: an antipsychotic (a dopamine-blocking medication) alongside psychological therapy and social support, with the Maudsley Prescribing Guidelines offering detailed dosing and switching advice. Where an organic cause is suspected, such as delirium or dementia with Lewy bodies, management prioritises treating the underlying trigger (for example correcting an infection or electrolyte disturbance), and antipsychotics are used cautiously or avoided altogether, since patients with dementia with Lewy bodies are highly sensitive to antipsychotic side effects and can develop severe, sometimes irreversible, reactions. Where a substance is responsible, such as alcohol withdrawal, management follows a structured withdrawal protocol, typically with benzodiazepines, guided by a CIWA (Clinical Institute Withdrawal Assessment) score. Charles Bonnet syndrome needs no antipsychotic at all: management is reassurance and optimising vision, because there is no underlying psychiatric illness to treat.

Consider an 80-year-old man brought in by his daughter with two weeks of vivid, detailed visual hallucinations of small children in his garden, alongside fluctuating alertness and new rigidity. The visual modality, the vivid detail, and the fluctuating consciousness together point strongly toward dementia with Lewy bodies rather than a primary psychotic illness, and this diagnosis should prompt real caution before starting any antipsychotic. The exam pearl to hold onto: visual hallucinations are organic until proven otherwise, and a hallucination with retained insight is a pseudohallucination, not a symptom of psychosis: examiners reward candidates who catch both traps rather than reflexively reaching for "schizophrenia."

Modality plus context, not the mere presence of a hallucination, drives the differential diagnosis, and insight is what separates a true hallucination from a pseudohallucination.