Article Text

Download PDFPDF

Hallucinations and illusions in migraine in children and the Alice in Wonderland Syndrome
  1. R A Smith1,
  2. B Wright2,
  3. Sophie Bennett2
  1. 1Department of Paediatrics, York Hospital, York, UK
  2. 2Adolescent Health and Child Psychiatry, Leeds and York Partnership NHS Foundation Trust, York, UK
  1. Correspondence to Dr R A Smith, Department of Paediatrics, York Hospital, Wigginton Rd, York YO31 8HE, UK; robertasmith246{at}


Design A prospective observational study over 1 year.

Setting A District General Hospital, and Child and Adolescent Mental Health Department.

Patients Children aged 8–18 years living in the catchment area of a district hospital service with any type of unexplained hallucinations or illusions associated with or without an established diagnosis of migraine.

Results The study identified nine children with a combination of migraine and a variety of hallucinations and illusions, including illusions of size, time, colour, body shape, movement and visual and auditory hallucination. An average of 10 symptoms (range 7–15) were reported.

Interventions None.

Main outcome measure None.

Conclusions It is important to recognise these symptoms to enable appropriate history taking and diagnosis. These symptoms are common and currently seem to go unrecognised and may pose diagnostic difficulties if onset is before typical migraine headaches occur.

  • migraine
  • hallucinations

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

  • There are case reports of perceptual illusions and hallucinations associated with migraine in children, but no previous prospective studies.

  • Migraine is common in children and can be associated with a variety of symptoms.

What this study adds

  • Hallucinations and illusions in children with migraine are more common than previously realised, may be present before headaches occur and may go unrecognised.

  • Children with migraine do not always report all their symptoms.

  • There are a wide range of reported symptoms and knowledge of these is important in making a diagnosis.

  • This study highlights a deficiency in International Classification of Headache Disorders 2nd edition.


A hallucination is a sensory perception in the absence of an external stimulus that is experienced as real. An illusion is a mistaken or false interpretation of a real sensory experience. Both can occur in migraine.1 Distinguishing between the two can be difficult in young children. Weber et al2 described 12 children with visual hallucinations and illusions in a retrospective study. Nine had a disturbed perception of size, five of form and four had irregular perception of movements. Seven (58%) had headaches that were thought to be migrainous.

There have also been case reports of auditory hallucinations without other psychiatric symptomatology during migrainous headaches. The hearing of voices has been reported as a migraine aura in patients without psychotic illness in adults and in children.

Dooley et al3 reported acute episodes with distortions in perception of time, associated movement and speech. The strong history of migraine and similar time course to an aura suggested a migrainous origin. Other case reports describe various other symptoms in association with migraine, including achromatopsia, prosopagnosia, alexia, metamorphosia, micropsia and macropsia, macrosomatognosia and microsomatognosia, depersonalisation, derealisation, illusions of time and a dreamy confused state. Olfactory, gustatory and other more complex hallucinations have also been reported in migraine.

The aim of the study was to further define the clinical association between migraine and of illusions or hallucinations in children and young people in a District General Hospital over the course of a year.


We included children aged 8–18 years living in the catchment area of a district hospital service with any type of unexplained hallucinations or illusions associated with or without an established diagnosis of migraine. We excluded those with a diagnosis of psychiatric disorder including psychosis, epilepsy, acute drug ingestion, acute febrile illness or significant hearing or visual impairment. The International Headache Classification system version II (IHCII)4 was used for defining migraine cases.

A letter was sent at the beginning of the study, explaining the study and inviting them to nominate possible participants, to all paediatricians and child and adolescent mental health staff at a District Hospital Foundation Trust and locality Child and Adolescent Mental Health Service (CAMHS) covering a population of approximately 300 000, with regular reminders throughout the study over a period of 1 year.

A letter and an information leaflet were sent to the parents and children identified as possible participants. Fully informed consent was obtained from those wishing to participate. After comprehensive case-note review, all participants were interviewed by RAS using a semistructured questionnaire. The questionnaire covered all previously reported types of hallucination and illusion. This was supported by a semistructured interview, the Present State Examination and the Spence Anxiety Scale to screen for anxiety disorders and psychosis. The study ran for 1 year (1 February 2011–31 January 2012).


Sixteen persons were invited to participate. Three declined and four did not respond to two invitations. Nine gave fully informed consent. The mean age was 12.7 years (range 9–16 years), five were female. The data from the study are presented in table 1.

Table 1

Symptoms reported from questionnaire based study

All the children fulfilled IHC II4 diagnostic criteria for migraine. There were no clinical diagnoses of epilepsy. All children either had a family history of migraine or a parent with recurrent severe headaches. Two parents had experienced perceptual distortions and illusions in association with episodes of migraine.

The mean age of onset of illusions or hallucinations was 8.4 years (range 5–16 years) and the mean age at onset of headaches was 9.4 years (range 7–15 years). Four children had onset of illusions or hallucinations before headaches, two had onset of symptoms at the same time as onset of headaches and three had onset of symptoms after onset of headaches. One child had abdominal migraine symptoms and had some perceptual illusions in association with the abdominal migraine episodes. In four children, the initial referral was because of illusions or hallucinations and in five children, because of headaches.

On some occasions, the illusion would last as long as the migraine episode; in others, they lasted only minutes. The frequency varied from several times per day to infrequent and sporadic. Some visual phenomena were associated with migraine episodes, some not, and in most it was variable. A wide range of illusions and hallucinations were reported and all had at least seven different symptoms. Many had not mentioned them during previous clinical consultations but were ascertained using the questionnaire. On questioning, this was because the children were reluctant to admit the symptoms; or because they did not consider them relevant when seen, with 78% reported seeing objects moving and becoming distorted. A high percentage had auditory hallucinations (78%). These were often the sound of voices that were distorted, for example, ‘as if a radio was on’ and not saying anything specific to the person. In two children, the voices were saying something in a voice with a sarcastic tone, but they were not specific about what was being said. One heard distant and sarcastic voices, one heard strange music and noises, another heard low-pitched mumbling male voices and one his father's voice. In none of the cases were the voices understandable or did they relay any specific message, information or instruction. Hyperacusis or phonophobia occurred in 55% of cases.

A non-specific and unpleasant feeling of fear was common (67%). Feelings of derealisation occurred in 55% and the feeling that the person was not real occurred in 44%. Visual hallucinations were common, occurring in 78% of children, and three involved seeing vague images of people. These were not recognisable people but were unclear images, with an associated unpleasant feeling of fear, while 55% experienced a sensation that parts of the body were changing in size. Distortions in perception of time were common and occurred in 67%, with a sensation of time speeding up in 55% and slowing down in 33%.

Two had high Spence anxiety scores within two SDs for the age-related scores, but at the high end of the normal range. Both felt that stress and worry exacerbated both their migraine headaches and perceptual experiences.

Two scored above the cut-off score for referral for assessment on the Present State Examination semi-structured interview. Both were fully assessed by CAMHS and psychosis or serious illness excluded. Psychiatric illness was not implicated in any of the young people.


This study demonstrates that illusions and hallucinations occur in children with migraine and may be more common than is currently recognised. These are different in nature to the type of auditory hallucinations seen in psychotic illness. First, there were no other accompanying psychotic symptoms and second, there was sometimes a temporal relationship with headaches. Children with hallucinations due to psychosis usually have additional features such as disturbances in language or thought, evidence of decreased motor activity, incongruous mood, bizarre behaviour and social withdrawal. In four of the nine cases, the illusions and hallucinations occurred before the onset of recurrent headaches and thus they can create a diagnostic challenge. It is important to be aware also that hallucinations in children can also be associated with anxiety, depression, psychosocial adversity, narcolepsy, hypnagogic and hypnopompic states, epilepsy, delirium, anaemia, post-traumatic stress, Tourettes syndrome, substance abuse and anaemia.

The hallucinations and illusions in some children cause anxiety (two in our sample) with some young people having difficulty understanding why they were experiencing them. Whether the anxiety is the cause or a consequence of the hallucinations and illusions is unclear. Auditory hallucinations have previously been considered rare in migraine. The more typical auditory symptoms experienced by migraineurs, such as unilateral tinnitus and hyperacusis are perceptual distortions postulated as being related to vasospasm of the vestibular or cochlear branches of the internal auditory artery.

Alice in Wonderland Syndrome (AIWS) was coined by Todd5 and refers to the book ‘Alice's Adventures in Wonderland’ published in 1865 by Charles Dodgson under the pseudonym Lewis Carroll, who may himself have had migraine.6 His book describes various illusions and hallucinations where Alice perceived herself to be smaller (microsomatognosia), larger (macrosomatognosia), changing shape (metamorphopsia) or objects around changing in size (micropsia and macropsia). Some of these co-occurred with depersonalisation, derealisation, visual illusions and illusions of time.

Most of the literature on AIWS is reported in case series, and there are no prospective or detailed clinical or population-based studies. Golden7 described two children where illusions of body image, time and vision occur. Dooley et al3 considered illusions of time to be more frequent than the more complex metamorphosias attributed to AIWS and proposed that the term ‘the rushes’ be used for perceptions of sounds or movements rushing by too quickly. Lewis Carroll mentions illusions of time in his book, ‘Through the Looking Glass’, in which he describes symptoms of speed without achieving distances.

There are International Classification of Headache Disorders 2nd edition (ICDH II)4 criteria for ‘childhood periodic syndromes that are commonly migraine precursors’ for cyclical vomiting, abdominal migraine and benign paroxysmal vertigo. While AIWS has been reported in association with migraine, glandular fever and some other febrile illnesses, the ICDH II4 does not recognise this subgroup of migraine. The symptoms described in patients with AIWS cannot, therefore, currently be classified according to these criteria as typical aura associated with migraine. We suggest that further research should be carried out to rectify this omission.

We have demonstrated that in a general paediatric department the only children referred to the study with a history of illusions or hallucinations (given the exclusion criteria stated) were those with a diagnosis of migraine. Many yielded much more information about their experiences and headaches upon fuller prompted history-taking. It is, therefore, vital that the nature and context of any reported hallucinations or illusions are interrogated in detail in association with full assessment of the possible psychiatric or other medical causes and comorbidities, in order to avoid a wrong diagnosis.

We recommend that this subgroup is captured in the International Classification of diseases and headaches and that future research is established to enable this.



  • Contributors RAS and BW had the initial idea for the project. RAS, BW and SB were involved in the design of the study. RAS and SB performed the literature searches. RAS interviewed all the families. RAS, BW and SB were involved in the writing and final version of the manuscript.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval National Research Ethics Service, Yorkshire and Humber Committee (REC 10/H1304/35).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles

  • Atoms
    R Mark Beattie