When they occur in the absence of any associated pathological abnormalities-with the exception of hearing impairment (hypoacusis)-they are called idiopathic musical hallucinations. Musical hallucinations are commonly divided into two groups. In a large sample ( n = 1007) of patients attending an audiology clinic, for example, auditory hallucinations were common in the last four weeks ( n = 114), with musical hallucinations ( n = 52) increasing in frequency with increasing hearing loss, and correlating with tinnitus and female sex. However, several studies in designated populations suggest that they are probably merely underreported. First described by Baillarger in 1844, musical hallucinations have traditionally been considered rare. They are different from earworms or ‘tunes in the head’, in the sense that they are perceptual in nature, whereas earworms are not. Musical hallucinations are auditory hallucinations characterised by songs, tunes, melodies, harmonics, rhythms, and/or timbres. Well over a century later, the relation between epilepsy and musical hallucinations still puzzles us. Ormerod quoted Schwartze, who had said that, ‘subjective aural sensations, which are caused by demonstrable affections of the ear, may in predisposed persons (.) become the direct cause of aural hallucinations’. The deafness progressed with increasing tinnitus, and gradually the woman also began to hear words, names, street sounds, and songs. In this pioneering paper, he described a 61-year-old woman with longstanding deafness and epilepsy.

In 1883, Ormerod wrote an article in Brain, entitled On epilepsy, in its relation to ear-disease, in which he mused that, ‘it needs no very extended observation to show that these two phenomena frequently co-exist’. We, therefore, conclude that musical hallucinations in epilepsy open a window to understanding these phenomena in a variety of conditions.īetween musical hallucinations and epilepsy a connection exists that was already suspected when neurology was still struggling to comprehend either phenomenon. An analysis of phenomenological characteristics, pharmacological triggers, and treatment effects suggests wider ramifications for understanding musical hallucinations. To account for the underlying mechanism we propose a Bayesian model involving top-down and bottom-up prediction errors within the auditory network that incorporates findings from EEG and MEG studies. The case descriptions suggest that musical hallucinations in epilepsy can be conceptualised as lying on a continuum with other auditory hallucinations, including verbal auditory hallucinations, and-notably-tinnitus. Based on the phenomenological descriptions and neurophysiological data, we distinguish four subgroups of epilepsy-related musical hallucination, comprising auras/ictal, inter-ictal and post-ictal phenomena, and phenomena related to brain stimulation. We also describe six of our own patients.

Our search yielded 191 articles, which together describe 983 unique patients, with 24 detailed descriptions of musical hallucinations related to epilepsy.

We, therefore, reviewed the literature, and assessed all descriptions of musical hallucinations attributed to epileptic activity. Their relation with epilepsy was first described over a century ago, but never systematically explored. Musical hallucinations are poorly understood phenomena.

Tinnitus and hallucinations