Accepted Manuscript

Title: Musical obsessions: A comprehensive review of

neglected clinical phenomena
Author: Steven Taylor Dean McKay Euripedes C. Miguel
Maria Alice De Mathis Chittaranjan Andrade Niraj Ahuja
Debbie Sookman Jun Soo Kwon Min Jung Huh Bradley C.
Riemann Jean Cottraux Kieron OConnor Lisa R. Hale
Jonathan S. Abramowitz Leonardo F. Fontenelle Eric A.
Storch
PII:
DOI:
Reference:

S0887-6185(14)00085-1
http://dx.doi.org/doi:10.1016/j.janxdis.2014.06.003
ANXDIS 1613

To appear in:

Journal of Anxiety Disorders

Received date:
Revised date:
Accepted date:

14-11-2013
9-5-2014
6-6-2014

Please cite this article as: Taylor, S., McKay, D., Miguel, E. C., Mathis, M. A. D.,
Andrade, C., Ahuja, N., Sookman, D., Kwon, J. S., Huh, M. J., Riemann, B. C.,
Cottraux, J., OConnor, K., Hale, L. R., Abramowitz, J. S., Fontenelle, L. F., and Storch,
E. A.,Musical obsessions: A comprehensive review of neglected clinical phenomena,
Journal of Anxiety Disorders (2014), http://dx.doi.org/10.1016/j.janxdis.2014.06.003
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

*Highlights (for review)

Highlights:

This is the first comprehensive review of musical obsessions.

It was based on the largest sample of published and unpublished cases ever

ip
t

assembled.

Phenomenology, differential diagnosis, comorbidity, and treatments are reviewed.

Musical obsessions differ from other obsessions in phenomenology and treatment.

Etiological hypotheses and important directions for future research are presented.

Ac
ce
p

te

an

us

cr

Page 1 of 60

*Manuscript
Click here to view linked References

1
Running head: Musical Obsessions

ip
t

Musical Obsessions: A Comprehensive Review of Neglected Clinical Phenomena

Steven Taylor a *, Dean McKay b, Euripedes C. Miguel c, Maria Alice De Mathis c,

cr

Chittaranjan Andrade d, Niraj Ahuja e, Debbie Sookman f, Jun Soo Kwon g, Min Jung

us

Huh g, Bradley C. Riemann h, Jean Cottraux i, Kieron OConnor j, Lisa R. Hale k,

Jonathan S. Abramowitz l, Leonardo F. Fontenelle m, & Eric A. Storch h, n
University of British Columbia, BC, Canada

Fordham University, NY, USA

University of So Paulo Medical School, Brazil

National Institute of Mental Health and Neurosciences, India

Northumberland, Tyne and Wear NHS Foundation Trust, UK

McGill University, PQ, Canada

Seoul National University, Republic of Korea

Rogers Memorial Hospital, WI, USA

University Lyon 1, France

Centre de Recherche, University Institute of Mental Health at Montreal, PQ, Canada

Kansas City Center for Anxiety Treatment, KS, USA

University of North Carolina at Chapel Hill, NC, USA

te

Ac
ce
p

m
n

an

Federal University of Rio de Janeiro, Brazil

University of South Florida, FL, USA

* Corresponding author: Department of Psychiatry, 2255 Wesbrook Mall, University of

British Columbia, Vancouver, BC, Canada, V6T 2A1. Email: taylor@mail.ubc.ca, Tel:
011-1-604-785-7558, Fax: 011-1-604-822-7756.
Supplementary material: Available at the journals website.

Page 2 of 60

2
Abstract
Intrusive musical imagery (IMI) consists of involuntarily recalled, short, looping
fragments of melodies. Musical obsessions are distressing, impairing forms of IMI that

ip
t

merit investigation in their own right and, more generally, research into these phenomena
may broaden our understanding of obsessive-compulsive disorder (OCD), which is

cr

phenomenologically and etiologically heterogeneous. We present the first comprehensive

us

review of musical obsessions, based on the largest set of case descriptions ever assembled
(N = 96). Characteristics of musical obsessions are described and compared with normal

an

IMI, musical hallucinations, and visual obsessional imagery. Assessment, differential

diagnosis, comorbidity, etiologic hypotheses, and treatments are described. Musical

obsessions may be under-diagnosed because they are not adequately assessed by current

measures of OCD. Musical obsessions have been misdiagnosed as psychotic phenomena,

te

which has led to ineffective treatment. Accurate diagnosis is important for appropriate
treatment. Musical obsessions may respond to treatments that are not recommended for

Ac
ce
p

prototypic OCD symptoms.

Page 3 of 60

3
Keywords: Obsessive-compulsive disorder, intrusive musical imagery, involuntary

Highlights:

ip
t

musical imagery, earworms, orhwurms, musical obsessions.

This is the first comprehensive review of musical obsessions.

It was based on the largest sample of published and unpublished cases ever

cr

us

assembled.

Phenomenology, differential diagnosis, comorbidity, and treatments are reviewed.

Musical obsessions differ in several ways from other obsessions regarding

an

Etiological hypotheses and important directions for future research are presented.

Ac
ce
p

te

phenomenology and treatment.

Page 4 of 60

4
1. Introduction
Obsessions and compulsions are the cardinal features of obsessive-compulsive
disorder (OCD). Obsessions are unwanted, intrusive, and distressing thoughts, images, or

ip
t

urges. Compulsions are repetitive behaviors or cognitive rituals that the person feels

compelled to perform, often in response to obsessions (American Psychiatric Association

cr

[APA], 2013). Although intrusive mental imagery has long been recognized as a salient

us

feature of OCD (de Silva, 1986), clinical descriptions and research have focused almost
exclusively on visual imagery. Phenomena involving intrusive musical imagery (IMI),

an

consisting of intrusive recollections of fragments of music (i.e., music running through

ones mind), have been neglected by contemporary OCD researchers. This is surprising

because descriptions of IMI appeared in the psychological and psychiatric literature for

over a century (e.g., Ebbinghaus, 1885; Janet, 1903; Kraepelin, 1915), and IMI continues

te

to be studied by contemporary musicologists and neurologists (Liikkanen, 2012b; Sacks,

2007). IMI is almost universally experienced (Liikkanen, 2012b). Underscoring the

Ac
ce
p

importance of IMI, the neurologist Oliver Sacks (2007) observed that it is the clearest
sign of the overwhelming, and at times helpless, sensitivity of our brains to music (p.
49). Musical obsessions are distressing, persistent, and disruptive forms of IMI.
OCD is etiologically heterogeneous, shaped by etiologic factors (genetic and

environmental) that are specific to a given class of symptom (e.g., etiologic factors
specific to washing compulsions) and by broader etiologic factors influencing many
different types of obsessive-compulsive (OC) symptoms and other forms of
psychopathology (Miguel et al., 2005; Taylor, 2011). Given the heterogeneity and
complexity of OCD, it is insufficient to study only the most common types of symptoms,

Page 5 of 60

5
such as washing or checking compulsions. A comprehensive understanding requires that
even the less common symptoms be investigated in order to gain a better understanding
of OCD.

ip
t

Although there have been few studies of musical obsessions compared to studies

of other OC phenomena, this is not necessarily an indication of the clinical insignificance

cr

of musical obsessions. Historically, clinical investigators have tended to underestimate

us

the prevalence and importance of OC-related phenomena. OCD was once considered
rare. It was not until the epidemiologic surveys were conducted that OCD came to be

an

recognized as a hidden epidemic (Jenike, 1989). Similarly, the prevalence and

significance of excessive hoardingan OC-related phenomenonwas not fully

appreciated until recently (Mueller, Mitchell, Crosby, Glaesmer, & de Zwaan, 2009).

Even as late as 1987, descriptive case studies of hoarding were sufficiently rare as to

te

merit publication (Greenberg, 1987).

To our knowledge, the present paper is the first comprehensive review of musical

Ac
ce
p

obsessions, based on the largest ever assembled set of case descriptions (N = 96,
consisting of 33 previously published cases and our 63 new cases). In this paper we will
review the characteristics of IMI in general, and then consider musical obsessions in
particular, including an examination of clinical features, differential diagnosis,
comparisons with visual obsessional imagery, etiologic considerations, assessment, and
treatment. We also highlight important gaps in the empirical literature and suggest
avenues for further investigation. The present paper aims to provide the groundwork for
subsequent research, by synthesizing and describing what is currently known, based on
published and unpublished research.

Page 6 of 60

6
2. Intrusive Musical Imagery
IMI is characterized by a tune that repeatedly comes to mind without the person
consciously trying to recall it (Williamson et al., 2011). Also known as earworms,

ip
t

ohrwurms, or involuntary musical imagery, IMI is a form of intrusive, auditory imagery

occurring in the absence of neurological pathology or ear disease. In a survey of over

cr

5,000 participants, Stafford (2012) found that IMI usuallybut not invariablyconsists

us

of simple repetitive song fragments, such as a particular verse or hook that has an
ineffable catchy quality. IMI need not be limited to simple or rhythmically repetitive

an

songs; almost any song can be a source of IMI (Hyman et al., 2013). IMI involves a
fragment of music that the person has previously heard, such as a chorus or line of music,

equal to or less than the capacity of auditory short-term memory. That is, repetitions of

15-30 second segments of music, persisting like a looping soundtrack (Bailes, 2007;

te

Beaman & Williams, 2010; Liikkanen, 2012b,c). Repetitions of IMI may last from
minutes to hours (Halpern & Bartlett, 2011), but most often consist of recurrent,

Ac
ce
p

intermittent episodes rather than a continuous musical soundtrack (Hyman et al., 2013).
IMI typically shifts over time; imagery of one particular tune might persist for a day or
two and then, days or weeks later, some other tune might become the new source of IMI
(Beaman & Williams, 2010; Hyman et al., 2013).
Episodes of IMI can be triggered by exposure to music, including repeated

exposure to a given song or jingle. To illustrate, Halpern and Bartlett (2011) found that
Christmas carols were most likely to become a source of IMI during the festive season,
when such music is most likely to be heard. IMI can also be triggered by stimuli
associated with particular pieces of music, such as by reading written lyrics or exposure

Page 7 of 60

7
to people or places associated with a particular piece of music (Halpern & Bartlett, 2011;
Hyman et al., 2013; Liikkanen, 2012a; Williamson et al., 2011). IMI can also be triggered
by personally important events (Janata, 2009; Reik, 1953). For example, music played at

ip
t

a funeral might become a source of later IMI.

Surveys indicate that more than 85% of people in the general population

cr

experience involuntary musical imagery at least weekly (Bailes, 2007; Bennett, 2002;

us

Liikkanen, 2012b). The frequency of IMI is positively correlated with the amount of
music involvement (e.g., amount of practice or listening to music), and the perceived

& Mllensiefen, 2012; Liikkanen, 2012b).

an

importance of music (Beaman & Williams, 2010; Bennett, 2002; Floridou, Williamson,

For most people, IMI is not aversive (Beaman & Williams, 2010; Halpern &

Bartlett, 2011; Hyman et al., 2013). In fact, some songs and advertising jingles become

te

popular precisely because they induce IMI. People who find IMI to be aversive are more
likely to have prototypic OC symptoms, such as washing or checking rituals. To

Ac
ce
p

illustrate, Williamson and Mllensiefen (2012) administered measures of IMI and OC

symptoms to a community sample of 1,024 people. OC symptoms were measured by the
total score on the Obsessive-Compulsive Inventory - Revised (a measure of common,
non-musical OC symptoms such as washing, checking, and ordering; Foa et al., 2002).
OC symptoms and the degree of distress associated with IMI were significantly related to
one another in causal modeling analyses in which OC symptoms were assumed to cause
IMI-related distress ( = .31). The rationale for assuming a causal relationship is unclear.
The authors failed to explain why one set of symptoms (OC symptoms) should cause
other symptom (distressing IMIs). It could be that the two were correlated because they

Page 8 of 60

8
were caused by a third variable, such as a vulnerability factor for OC symptoms (see
below for etiologic considerations). Nevertheless, the findings show that prototypic OC
symptoms and IMI-related distress are correlated with one another.

ip
t

3. Musical Obsessions: Descriptive Psychopathology

3.1. Diagnostic Criteria

cr

Musical obsessions are not mentioned in the most recent version of the Diagnostic

us

and Statistical Manual of Mental Disorders (DSM-5; APA, 2013). Musical obsessions are
described in the clinical literature as episodes of IMI that meet criteria for OC symptoms;

an

that is, the obsessional imagery is recurrent, persistent, intrusive, unintentional, time
consuming (i.e., more than an hour per day), and causes distress or functional

impairment. Note that this definition refers to persistent, involuntary musical imagery. It

does not refer to obsessional preoccupations about particular pieces of music. A woman

te

with an obsessional fear of becoming a lesbian, for example, might fear and avoid songs
with homoerotic lyrics, such as Katy Perrys I Kissed a Girl and I Liked it. This is not a

Ac
ce
p

musical obsession as the term is used. Similarly, the definition of musical obsessions
excludes cases in which the person feels compelled to repeatedly listen to, or recall a
piece of music until some perfectionistic just right sense of completeness has been
attained. Musical obsessions are also distinct from sought-out musical recollections,
which have been described by composers who are absorbed in thinking about and writing
music, and by some music enthusiasts. A person who ardently collects and repeatedly
listens, with enjoyment and without impairment, to the many different versions of
Beatles songs can be considered to have a musical preoccupation, not a musical
obsession as the term is defined.

Page 9 of 60

9
3.2. Differential Diagnosis
3.2.1. Musical hallucinations. Hallucinations are perceptions lacking an adequate
input stimulus, which are perceived to arise from an external source and are interpreted as

ip
t

veridical (Sedman, 1966). Musical hallucinations are rare (Berrios, 1990). They have

been described in patients with acquired hearing loss, epilepsy, structural brain lesions,

cr

psychosis, delirium, drug induced states, and hypnagogic states (Evers & Ellger, 2004).

us

In contrast to musical hallucinations, musical obsessions are a form of imagery that the
person recognizes as originating from his or her mind, and occur in people who do not

an

suffer from hearing disorders or neurological pathology.

3.2.2. Pseudohallucinations. The concept of pseudohallucinations is problematic

because of the lack of consensus regarding its definition and distinction from true

hallucinations (Sanati, 2012). However, it is generally agreed that pseudohallucinations

te

are forms of imagery that occur in inner (subjective) space and are recognized by the
person as not veridical perceptions. To this extent, musical pseudohallucinations

Ac
ce
p

resemble musical obsessions. The two differ in that pseudohallucinations need not be
perceived as aversive and are not associated with compulsions (Oyebode, 2001). Unlike
musical obsessions, which are not due to drug intoxication or deafness,
pseudohallucinations can occur exclusively during hallucinogen intoxication and can
arise as a result of severe hearing loss (El-Mallakh & Walker, 2010).
Pseudohallucinations can occur in psychotic disorders such as schizophrenia, and in
various non-psychotic disorders such as dissociative disorders and borderline personality
disorder (Oyebode, 2001; Sanati, 2012). Musical obsessions and pseudohallucinations
also differ in terms of insight. By definition, insight is always present in musical

Page 10 of 60

10
obsessions; that is, the person always recognizes that the musical imagery is a product of
his or her mind. In comparison, insight can fluctuate in musical pseudohallucinations; at
times the person might believe that the source of the music is internal, but at times may

ip
t

believe that it comes from an external source (Sanati, 2012).

3.2.3. Palinacousis. This is a neurological symptom consisting of episodes of

cr

paroxysmal auditory hallucination, in which perseveration (echoing) of an external

us

auditory stimulus occurs after cessation of the stimulus. The person recognizes the
hallucinatory nature of the experience, which is often described as hearing a fragment

an

of music, a noise, or portions of speech (Jacobs, Feldman, Diamond, & Bender, 1973;
Mohamed, Ahuja, & Shah, 2012). Palinacousis is rare and musical palinacousis is even

rarer (Mohamed et al., 2012; Podoll, 2010). Palinacousis is associated with EEG and

neuroimaging abnormalities, with structural brain pathology (e.g., cortical atrophy,

te

primary tumors, and metastases), typically in the temporal or parietal lobes (Mohamed et
al., 2012). Palinacousis is usually a manifestation of seizure activity (e.g., an aura),

Ac
ce
p

although it may also be a post-ictal or a migrainous phenomenon (Di Dio et al., 2007;
Mohamed et al., 2012; Podoll, 2010). Palinacousis is associated with marked
neurological deficits (e.g., aphasia; Di Dio et al., 2007), whereas musical obsessions are
not associated with such pathology. Unlike musical obsessions, which occur in ones
mind with no auditory localization to a given ear, the sounds in palinacousis often occur
only in one ear (Mohamed et al., 2012). Musical obsessions also differ from palinacousis
in that the former are limited to repetitions of fragments of music the person has heard in
the past, rather than being simply recollections of the last thing the person heard.
3.3. Characteristics of Musical Obsessions: Findings from 96 Cases

Page 11 of 60

11
3.3.1. Acquisition of published case material. A literature search of the following
databases was conducted up to May 1, 2014: PsychInfo, MEDLINE, EMBASE, and
Google Scholar. The keywords were obsess* and music* (the asterisks denote the use of

ip
t

wild cards). Articles in all languages were considered. References in the identified

articles were examined for additional relevant material. Researchers who had previously

cr

published on musical obsessions were also contacted for published and unpublished

us

cases. A total of 17 publications were identified (Ahuja, 2001; Akhtar, Wig, Varma,
Pershad, & Verma, 1975; Andrade & Rao, 1997; Berg, 1953; Cameron & Wasielewski,

an

1990; Fontenelle, 2008; Gomibuchi, Gomibuchi, Akiyama, Tsuda, & Hayakawa, 2000;
Hemming & Altenmller, 2012; Marani, Aukst-Margeti, Grgi, & Kumi, 2011;

Matsui et al., 2003; Mendhekar & Andrade, 2009; Nath, Bhattacharya, Hazarika, Roy, &

Praharaj, 2013; Pfizer & Andrade, 1999; Praharaj et al., 2009; Rapoport, 1980; Saha,

te

2012; Zungu-Dirwayi et al., 1999). The five cases reported by Hemming and Altenmller
(2012) were excluded because most suffered from tinnitus.

Ac
ce
p

Berg (1953) summarized findings from 10 cases, Gomibuchi et al. (2000)

presented illustrative information on two of their five cases, four other studies each
presented two cases (Ahuja, 2001; Matsui et al., 2003; Rapoport, 1980; Zungu-Dirwayi et
al., 1999), and the remaining articles each described a single case. Thus, we were able to
locate published reports of 33 cases of musical obsessions. We also identified a study
comparing diagnostic groups in terms of the prevalence of musical obsessions (Hermesh
et al., 2004). Those authors equated musical hallucinations with musical obsessions. That
study was not included in our count of cases of musical obsessions because of Hermesh
et al.s questionable decision to equate musical hallucinations with musical obsessions.

Page 12 of 60

12
3.3.2. Additional cases. To supplement the published case studies, the authors of
the present paperclinicians with expertise in OCDreviewed their own clinical case
files for previously unpublished cases of musical obsessions. All cases had been

ip
t

evaluated with Yale-Brown Obsessive-Compulsive Scale (Y-BOCS; Goodman et al.,

1989). The patients had been evaluated in a variety of settings, ranging from private

cr

practices focusing on the treatment of anxiety and related problems to large-scale hospital

us

or research clinics specializing in OCD and related problems. The number of patient files
that were examined ranged from a few dozen to databases consisting of over a thousand

an

clinical records. We obtained information on 63 unpublished cases. Thus, we had

information on a total of 96 previously published and new cases of musical obsessions.

To the best of our knowledge, this is by far the largest set of cases of musical obsessions

ever assembled.

te

The representativeness of the published and unpublished cases cannot be

determined on the basis of the available information; that would require epidemiologic

Ac
ce
p

research, which has yet to be conducted. Nevertheless the published and new cases
provide the most comprehensive collection of clinical reports to date and, in aggregate,
provide a preliminary portrait of the nature and treatment of musical obsessions. Details
of the cases appear in the Appendix. The following sections summarize the main
findings.

3.3.2. Demographic features. The 96 cases were obtained from throughout the
world: Brazil, Canada, India, Japan, Republic of Korea, South Africa, and the U.S. The
mean age was 33 years (range 9-83 years) and 50% were female. About half (53%) were
employed, with the remainder being students (29%), homemakers (10%), retired (4%), or

Page 13 of 60

13
unemployed (4%).
3.3.3. Diagnostic features. With the exception of Berg (1953), for which
diagnostic status was unclear, all cases were diagnosed with OCD. There was no

ip
t

evidence that any of these individuals suffered from palinacousis (e.g., none had a history
of seizures) and none appeared to be suffering from tinnitus or deafness. Musical

cr

obsessions of two patients were initially misdiagnosed as symptoms of schizophrenia.

us

However, neither they nor any of the other cases had psychotic features such as formal
thought disorder or verbal auditory hallucinations. Neurological factors (brain

an

inflammation or injury) might have contributed to three cases of musical obsessions,

evaluations (e.g., EEG, neuroimaging).

discussed below, but even these cases had unremarkable findings on neurological

3.3.4. Content of obsessions. Musical obsessions were either the sole presenting

te

problem or occurred in the context of other psychopathology. Musical obsessions were

either primary or secondary to other presenting problems in terms of severity. Musical

Ac
ce
p

obsessions consisted of either a single song or a series of songs that changed over time, as
in the case with IMI in general. All musical obsessions were songs or melodies that the
person had heard recently or in the past, and were often popular songs. In some cases,
however, the songs were more specific, sometimes with a special meaning to the
individual. For one person, a professional musician, the musical obsessions consisted of
classical music he had played (e.g., Ravels Bolro) or music he had recently composed.
In other cases, the obsessions consisted of advertising jingles or pieces of film scores. In
one case the obsession consisted of cellular phone ringtones. The duration of musical
obsessions generally ranged from months to years. In some cases the obsessions occurred

Page 14 of 60

14
intermittently and in other cases were reported as occurring almost continuously during
waking hours.
3.3.5. Fear, avoidance, and compulsions related to musical obsessions. It is

ip
t

common for people with OCD, regardless of the type of symptoms, to attempt to cope

with their obsessions by engaging in avoidance behaviors or some form of compensatory

cr

ritual (Abramowitz et al., 2011). For example, someone with visual obsessional imagery

us

of harming their child might avoid kitchen knives, seek reassurance from others about the
childs well-being, and attempt to suppress or eliminate the unwanted violent images. We

an

found evidence of analogous behaviors in the cases of musical obsessions.

In some cases of musical obsessions, patients developed fear and avoidance of

stimuli that triggered musical obsessions, including avoidance of stimuli or situations

associated with music, such as avoiding listening to music or avoiding public places

te

where music is commonly played. To illustrate, a patient with ring tone obsessions went
to great lengths to avoid exposure to ring tones. He wore earplugs, kept his phone on

Ac
ce
p

silent (vibrate) mode, urged others to do the same, and avoided places where he might
hear the ringing of cellular phones (Mendhekar & Andrade, 2009). In other cases,
patients worried about, and made unsuccessful attempts to suppress, the unwanted
musical imagery (Matsui et al., 2003). Thought suppression involved attempts to replace
the unwanted IMI with other, distracting stimuli (i.e., cognitive rituals). In at least one
case, the mere thought that one might hear unwanted music was sufficient to trigger the
IMI, leading the person to engage in efforts at avoidance or thought suppression.
3.3.6. Circumstances of onset. Onset in a number of cases occurred after the
person had been spending a lot of time listening to music. For example, prior to onset of

Page 15 of 60

15
ring tone obsessions the patient had enjoyed downloading, previewing, and changing ring
tones on his cellular phone, devoting up to 2-3 hours/day to this activity. He discontinued
the activity as he became increasingly troubled by recurrent unwanted intrusive imagery

ip
t

of ring tones.

The onset and exacerbation of musical obsessions in other cases appeared to be

cr

stress related. In several cases, musical obsessions developed in students who were

us

studying for exams. In another case, musical obsessions consisted of funeral dirges,
which developed shortly after the death of the patients wife. In yet another case the

an

obsessions began after the patient resigned from his job due to interpersonal conflicts. In
one case the musical obsession appeared to be the result of an attempt by the patient to

distract herself from other obsessions. She developed the strategy of singing to herself to

avoid or neutralize her other, non-musical obsessions, but then began experiencing

te

unwanted, distressing musical imagery of the songs she had used as distraction.
There was no evidence of neurological abnormalities in almost all cases. In three

Ac
ce
p

cases, however, onset might have been triggered by a neurological insult. In one case
(Cameron & Wasielewski, 1990), musical obsessions began a month after the patient
received a blow to the head. However, a comprehensive evaluation revealed no evidence
of neurological abnormalities or non-musical auditory echoing or seizures. The patient
was successfully treated with clomipramine. Although one should be cautious about
drawing conclusions about etiology based on treatment response, the patients response to
clomipramine was consistent with what is typically reported for OCD in general. There is
no evidence that a neurological condition such as palinacousis would respond to
clomipramine.

Page 16 of 60

16
In a second case (Pfizer & Andrade, 1999), musical obsessions arose suddenly, in
conjunction with an infection causing sore throat and fever. The infection was
successfully treated with erythromycin but the musical obsessions persisted. The clinical

ip
t

presentation resembled the class of pediatric autoimmune neuropsychiatric disorders

associated with streptococcal infection (PANDAS). PANDAS-related OCD is said to

cr

arise from infection-related inflammation of the basal ganglia and associated structures.

us

The validity of the concept of PANDAS is contentious (Murphy, Kurlan, & Leckman,
2010); Despite more than a decade of studying PANDAS, it is still not possible to

an

confirm its existence and whether it is a post-streptococcal autoimmune disorder

(Knupp, de Oliveira, & Pelajo, 2010, p. 103). Pfizer and Andrade (1999) concluded that

the association between infection and the onset of musical obsessions was probably

coincidental.

te

In a third case (Zungu-Dirwayi et al., 1999, Case 2), musical obsessions

developed six months after a closed head injury. The patient was unconscious for four

Ac
ce
p

days and had a month of retrograde amnesia. A CT scan was normal although there was
evidence of personality changes, in which the patient became apathetic with intermittent
aggressive outbursts. Zungu-Dirwayi et al. (1999) regarded the case as one of acquired
OCD.

3.3.7. Comorbidity. Comorbid OC symptoms (current or lifetime comorbidity)

were most extensively investigated in the unpublished cases, where patients completed
the Y-BOCS symptom checklist. The published cases, however, also provided some
information on comorbidity. Most patients suffering from musical obsessions had past or
current comorbid OC symptoms, including washing, checking, or ordering compulsions,

Page 17 of 60

17
and aggressive, horrific, sexual, or religious obsessions. Some of the comorbid
obsessions consisted of distressing visual imagery. Some patients had comorbid
obsessions that were similar in form to musical obsessions; for example, unwanted,

ip
t

intrusive recollections of verses of spoken poetry. Comorbidity of musical obsessions

with many other types of OC symptoms is consistent with OCD in general, in which such

cr

comorbidity is common (APA, 2013).

us

Several patients were described as having features of obsessive-compulsive

personality disorder (i.e., extreme perfectionism or preoccupation with orderliness). One

an

patient was described as suffering from distressing pareidolias, that is, illusory images
that were perceived while looking at shapes (e.g., seeing gorillas in the mosaic pattern of

floor tiles). Another patient exhibited misophonia, which is characterized by intolerance

and distress in response to minor everyday sounds such as the sound of other people

te

eating (Mller, 2011). Disorders that are commonly comorbid with OCD in general were
also described, such as other anxiety disorders, mood disorders, and tics. There was no

Ac
ce
p

evidence of psychotic features in the 96 cases. In summary, for most cases the pattern of
comorbidity was broadly similar to that of OCD in general.
3.3.8. Prevalence. There have been no epidemiologic studies that assessed

musical obsessions. It is widely assumed that musical obsessions are rare (Nath et al.,
2013; Rapoport, 1980; Saha, 2012). Prevalence may have been underestimated because
of a lack of clinical attention to the phenomena (i.e., clinicians failing to inquire about the
phenomena), because of limitations in our assessment methods, and patients often fail to
disclose the full range of their obsessions and compulsions unless they are specifically
asked about particular types of symptoms (Andrade & Rao, 1997; Marani et al., 2011).

Page 18 of 60

18
Current measures of OC symptoms and OCD fail to adequately assess musical
obsessions, which makes it difficult to estimate their prevalence. Only one item of the
symptom checklist of the Y-BOCS assesses whether the respondent has suffered from

ip
t

intrusive nonsense sounds, words, or music, with the example given as words, songs,
or music in your mind that you cant stop. This item fails to distinguish musical

cr

obsessions from other intrusive auditory imagery.

us

4. Musical Obsessions Compared to Obsessions

Characterized by Intrusive Visual Imagery

an

To further clarify the nature of musical obsessions, it is useful to consider how

they are similar to, and different from, other types of obsessional imagery. Most

obsessional imagery is visual (de Silva, 1986; Speckens et al., 2007). Obsessions

characterized by spoken words, olfactory, gustatory, or tactile imagery are rare (Chauhan,

te

Shah, & Grover, 2010; de Silva, 1986; Speckens et al., 2007). Unwanted visual images
are typically inherently aversive or violate the persons values or beliefs. Common

Ac
ce
p

examples include images of violently attacking ones child or elderly parents,

blaspheming, or performing sexually inappropriate acts. Other obsessional images relate
to fears of contamination, disease, or loss of mental control, or abhorrent events.
Examples of the latter include images of children being killed, mutilated and bloody
corpses, dead animals, cars colliding, parents having sexual intercourse, or seeing oneself
covered in excrement (e.g., Akhtar et al., 1975; de Silva, 1986; Emmelkamp & Kwee,
1977; Lipton, Brewin, Linke, & Halperin, 2010; Speckens et al., 2007).
Musical obsessions differ from typical visual obsessional imagery in that the
intrusive musical imagery is not inherently abhorrent or aversive, and does not involve

Page 19 of 60

19
ego-dystonic material such as unwanted images that violate the persons religious, sexual,
or moral values. The intrusive music is usually not something that the person associates
with some unpleasant event, although there are some exceptions (e.g., the funeral dirge

ip
t

case). Musical and visual obsessional imagery also differ in that the former involves
music that is typically familiar to the person; that is, something that the person has

cr

actually heard or composed. In comparison, visual obsessional imagery typically consists

us

of imagined scenarios. Musical and visual obsessional imagery also differ to some extent
in their treatment. Imaginal (and situational) exposure and response prevention is an

an

empirically supported treatment for OCD in general, including the treatment of visual
obsessional imagery (Abramowitz, Deacon, & Whiteside, 2011).These methods have

been used to treat musical obsessions, but distraction has also been successfully used in

case studies (see below). Distraction is not an empirically supported treatment for visual

te

obsessions, as discussed below. Musical and visual obsessional imagery are similar in
that both are commonly comorbid with other OC symptoms, such as washing, checking,

Ac
ce
p

and ordering compulsions. Our case series indicates that it is not uncommon for musical
obsessions to be comorbid with visual obsessions. The comorbidity and
phenomenological similarities and differences between musical and visual obsessional
imagery raise the question of whether there are similarities and differences in their
etiologic mechanisms. This remains to be investigated.
Obsessions involving visual imagery can be explained in terms of contemporary
cognitive-behavioral models of OCD (see below). Given that musical obsessions differ
from visual obsessional imagery in important ways, as discussed above, it seems likely
that current cognitive-behavioral models would require modification if they are to be

Page 20 of 60

20
extended to account for musical obsessions. In the following section we summarize these
models and discuss how they might be modified to explain why most people experience

5. Etiological Considerations
5.1. Contemporary Cognitive-Behavioral Models of OC Symptoms

ip
t

IMI but relatively few develop musical obsessions.

cr

Although there is no definitive etiological model of obsessions and compulsions,

us

contemporary cognitive-behavioral models are among the leading approaches (Frost &
Steketee, 2002; Salkovskis, 1996). These models, which share basic postulates, propose

an

that OC symptoms arise from specific types of dysfunctional beliefs, where the content
and strength of belief influences the development and severity of OC symptoms. Three

inter-correlated types of beliefs have been theoretically and empirically linked to OC

symptoms: (a) perfectionism (P) and intolerance of uncertainty (C; collectively referred

te

to as PC), (b) over-importance of thoughts and the need to control thoughts (ICT), and (c)
inflated responsibility and the overestimation of threat (RT) (Obsessive Compulsive

Ac
ce
p

Cognitions Working Group, 2005). PC involves beliefs that mistakes and imperfection
are intolerable, along with beliefs that it is necessary and possible to be completely
certain that aversive events will not occur. ICT entails beliefs that the mere presence of
unwanted thoughts indicates that such thoughts are important or portentous (e.g., the
belief that bad thoughts, even unwanted ones, lead to bad deeds), along with beliefs
that complete control over ones thoughts is necessary and possible. RT includes beliefs
that aversive events are quite likely to occur and that one has a duty to prevent such
events.
The presence of strongly held dysfunctional beliefs are said to play a role in

Page 21 of 60

21
transforming ordinary intrusions into obsessions. To illustrate, consider the unwanted
intrusive image of stabbing a loved one. If the person interprets the intrusion as having no
significance (i.e., regarded it as mental detritus) then he or she would experience little

ip
t

distress or impairment due to the unwanted image. In comparison, if a person interpreted

the intrusion as an indication that he or she is at risk for losing control and inflicting

cr

harm, then the person would become distressed, attempt to vigorously suppress such

us

imagery, avoid triggering stimuli (e.g., kitchen knives), and repeatedly seek reassurance
that loved ones have not been harmed (Abramowitz et al., 2011; Frost & Steketee, 2002).

an

Consistent with this formulation, research shows that the frequency or severity of
obsessions in general is predicted by the strength of a persons beliefs about the

significance of ones unwanted thoughts (e.g., Violent thoughts lead to dangerous

deeds), and by beliefs about the importance of controlling such thoughts (e.g., If I dont

te

drive the violent thoughts out of my mind, I will lose control and inflict harm on others)
(Taylor, Abramowitz, McKay, & Cuttler, 2012).

Ac
ce
p

5.2. Toward a Model of Musical Obsessions

The cognitive-behavioral formulation can be extended, with modification, to

musical obsessions. We propose a two-factor model for musical obsessions, consisting of

(a) individual difference variables influencing the frequency and persistence of normal
IMI, interacting with (b) dysfunctional beliefs about IMI. Two factors are proposed
because dysfunctional beliefs alone are unable to account for musical obsessions. If a
person rarely or never experiences normal IMI, then the opportunity to misinterpret the
meaning or significance of the intrusions does not arise. People who experience
unusually frequent and persistent IMI are more likely to worry about the meaning and

Page 22 of 60

22
consequences of their unwanted, uncontrollable musical imagery. Therefore, individual
differences in the frequency and duration of IMI seem likely to play some role in
influencing the development of musical obsessions. However, even highly frequent and

ip
t

persistent IMI is insufficient by itself to account for musical obsessions. If a person

interprets highly persistent IMI as an innocuous background soundtrack to daily life then

cr

IMI would not become a source of distress or lead to impairing avoidance behaviors or

us

compulsions. Negative beliefs and appraisals seem important for transforming inherently
innocuous IMI into musical obsessions.

an

Negative appraisals of IMI can arise from beliefs that IMI is abnormal (e.g., a
harbinger of mental deterioration), a sign that one lacks sufficient self-control (e.g., If I

cant control my mind, then I might lose control of my behaviors), or a sign of some

other aversive outcome (e.g., The music in my head interferes with my concentration; if

te

I cant concentrate, then I might fail at school). Threat-related misinterpretations may

lead to maladaptive coping responses that in turn increase the frequency of IMI. This

Ac
ce
p

pattern of maladaptive coping is suggested in the cases of musical obsessions reported by

Gomibuchi et al. (2000). The patients were perfectionistic students who considered it
important to study hard so that they could gain admittance to the best universities:

They tried hard to get rid of the music that, they felt, was abnormal. Their attempt,
therefore, focused more attention on the symptom [the intrusive music]. The more
they tried to get rid of it, the more they focused attention on it. This is the vicious
cycle ... Their efforts not to hear it, paradoxically, made it occur more frequently
and more intensively. Through this process, musical obsession was formed.

Page 23 of 60

23
(Gomibuchi et al., 2000, p. 206)

This account is similar to cognitive-behavioral accounts of OCD in general,

ip
t

concerning the paradoxical, self-defeating effects of attempting to suppress obsessional

thoughts (Salkovskis, 1996).

cr

5.3. What Causes Individual Differences in IMI?

us

Very little is known about the factors influencing individual differences in the
frequency or persistence of IMI. This is an important gap in the research literature.

an

Neural imaging and other forms of research suggest that auditory imagery involves the
same brain areas as those involved in auditory perception (Hubbard, 2010; Zatorre &

Halpern, 2005). These include the superior temporal gyrus, frontal and parietal lobes, and

supplementary motor cortex (Hubbard, 2010; Janata, 2009). Levitin (2006) speculated

te

that repetitive IMI occurs because the neural circuits representing the song get stuck in
playback mode (p. 155). Given the multitude of structures involved in auditory

Ac
ce
p

imagery, it is likely that the interconnecting circuitry is complex. To better understand

musical obsessions, further research is required to identify the factors that shape
individual differences in the involuntary activation and persistence of musical imagery.
Neuroimaging research can also be conducted in order to determine whether people with
low versus high frequency IMI differ in patterns of activation of brain structures
associated with auditory imagery.
Highly frequent IMI might arise from dysfunctions in mechanisms involved in the
inhibition and retrieval of auditory memories, causing memories to be too readily
triggered. If this is the case then patients with musical obsessions should also experience

Page 24 of 60

24
recurrent intrusions of verbal auditory imagery. There was suggestive evidence of this in
our case series. One patient had musical obsessions along with auditory verbal intrusions
consisting of the name of a dead friend that echoed in her mind (Cabral, Cabral,

ip
t

Cabral; Fontenelle, 2008). In two other cases the person experienced intrusions

consisting of lines of spoken poetry (Andrade & Rao, 1997; Saha, 2012). Further research

cr

is required to investigate whether musical obsessions are part of a general hyper-retrieval

6. Treatment

us

of auditory memories.

an

The available database of cases provides leads as to how musical obsessions

might be effectively treated, and provides a basis for developing treatment protocols that

could be empirically evaluated in future research. Details of the treatments for the cases

te

cases and the relevant literature.

appear in the Appendix. The following is a summary of the salient points arising from the

6.1. Assessment for Treatment Planning

Ac
ce
p

In assessing musical obsessions the clinician cannot rely solely on the Y-BOCS

because it does not specifically assess musical obsessions. The clinician needs to
specifically inquire about whether the patient suffers from musical obsessions and should
rule out differential diagnoses (see above). If cognitive-behavioral treatments are
planned, then it would be useful to assess the patients beliefs and appraisals of unwanted
musical imagery, to determine whether the person is making catastrophic
misinterpretations (e.g., The uncontrollable music in my head means that I have
something seriously wrong with my brain).
Unusual OCD symptoms can lead to misdiagnosis and inappropriate treatment

Page 25 of 60

25
(Marani et al., 2011). In our case series, two patients were misdiagnosed as having a
psychotic disorder (Marani et al., 2011; Nath et al., 2013). In neither case did the
patient benefit from antipsychotics; in fact, there was evidence of symptom worsening.

ip
t

When musical obsessions were correctly diagnosed, treatment was changed to

medications known to be efficacious for OCD (e.g., clomipramine, fluvoxamine) and the

us

obsessions is important for planning appropriate treatment.

cr

patients began to clinically improve. Accordingly, an accurate assessment of musical

6.2. Pharmacotherapy

an

For musical obsessions in general, including those that were correctly diagnosed
at the outset, our case studies suggest that patients tended not to benefit from

antipsychotic medications (e.g., risperidone, clozapine, haloperidol). Such drugs may be

effective in treating musical hallucinations (Evers & Ellger, 2004) and verbal auditory

te

hallucinations associated with schizophrenia (Martinez, Marangell, & Martinez (2008).

The case material also suggests that patients with musical obsessions are more likely to

Ac
ce
p

benefit from medications that are efficacious for OCD in general, such as selective
serotonin reuptake inhibitors (SSRIs) and clomipramine.
6.3. Cognitive-Behavioral Interventions
Cognitive restructuring may be useful to address threat-related beliefs about the

consequences of musical obsessions. Cognitive therapy can be used to target

dysfunctional beliefs such as The uncontrollable music in my head means that I wont
be able to function and will fail at school, or I will lose mental control. Several cases
appeared to benefit from acceptance-based interventions. These involved educating
patients about the normalcy of IMI and encouraging them to accept the intrusive music,

Page 26 of 60

26
without trying to suppress it. As had been reported for other obsessions, the effort the
person puts into trying to suppress IMI is correlated with the frequency of IMI
(Williamson & Mllensiefen, 2012). Reappraisal and acceptance strategies show promise

ip
t

for reducing distress and reduce maladaptive efforts at suppressing musical obsessions

(Gomibuchi et al., 2000). A helpful coping statement might be: This is just music in my

cr

head; there is no need to dwell on it or fight it because it will eventually pass.

us

6.4. Distraction

Distraction-based treatments are not recommended for prototypic OC symptoms

an

(Abramowitz et al., 2011). However, there are several lines of evidence suggesting that
they can be useful in treating musical obsessions. Studies of normal IMI suggest that

distraction can be useful. Retrospective surveys and prospective diary research indicates

that people most commonly use distraction to terminate episodes of normal IMI (e.g.,

te

keeping busy with whatever task is at hand, or listening to a competing piece of music;
Beaman & Williams, 2010). In an experimental analogue study of normal volunteers,

Ac
ce
p

Hyman et al. (2013) found that normal IMI was disrupted when the person engaged in
tasks producing a moderately high load on working memory (i.e., completing easy
anagrams or Sudoku puzzles).

Distraction can take the form of competing stimuli (such as competing sounds)

that are incompatible with forming melodies or incompatible with the specific intrusive
melody (such as other dissimilar songs). One of the authors (DM) successfully used a
distraction method in which the patient was asked to sing songs other than those
characterizing the musical obsession (see Appendix). The patient was also asked to sing
the target obsessional song aloud but incorrectly. In this approach musical obsessions

Page 27 of 60

27
were conceptualized as similar to depressive ruminations, which can be successfully
treated with distraction strategies (Teismann, Michalak, Willutzki, & Schulte, 2012). The
additional component involving participants singing aloud the target song, with the

ip
t

clinician, may weaken the specificity of the IMI. That is, rather than only experiencing
the obsession of the song as heard in its original form, the same melody now has

cr

associations with other sources. This is based on the assumption that the musical

us

obsession has a singular association (song = aversive experience). With additional

experiences of the melody under new conditions, the obsession has an expanded base of

an

associations (song = aversive experience; song = clinical practice in patient and

clinicians voice; song = other emotional experience). This expanded base is assumed to

lead to greater flexibility in associations, derived from a relational frame theory (Dymond

& Roche, 2013).

te

6.5. Exposure and Response Prevention (ERP)

The successful use of distraction or competing tasks in some cases of musical

Ac
ce
p

obsessions is quite different from treatments for other types of OC symptoms.

Empirically supported cognitive-behavioral treatments for (non-musical) OCD generally
eschew the use of distraction because it is ineffective and sometimes counter-productive
(Abramowitz et al., 2011). Effective treatment of obsessional thoughts or images involves
prolonged imaginal exposure to distressing material, while refraining from cognitive
compulsions (e.g., refraining from replacing a bad thought or image with a good one;
Abramowitz et al., 2011).
Some of our case reports of patients with musical obsessions suggest that both
distraction/competing task treatments can be effective, whereas our other case reports

Page 28 of 60

28
suggest that ERP can be effective. Case reports are useful in that they suggest areas for
further investigation. Future studies could directly compare distraction versus ERP as
treatments for musical obsessions, and to investigate whether there are any variables.

ip
t

6.6. Summary

There are several promising approaches to treating musical obsessions that merit

cr

further investigation in randomized, controlled studies. Musical obsessions may respond

us

to medications and to cognitive-behavioral protocols used to treat OCD in general.

However, some cases of musical obsessions have responded to treatments involving

an

distraction or stimulus substitution. ERP, cognitive interventions, SSRIs, and

clomipramine are first-line treatments for OCD in general, so it is reasonable to

implement these interventions singly or in combination in the treatment of musical

obsessions. If these interventions are unsuccessful then we recommend that the clinician

te

reevaluate the diagnosis to determine whether the patient has musical obsessions or some
other clinical condition such as a psychotic or neurological disorder. If the most likely

Ac
ce
p

diagnosis is musical obsessions, then distraction-based interventions could be considered.

7. Discussion

7.1. Conclusions

The present paper is, to our knowledge, the first comprehensive review of musical

obsessions, based on the largest set of case studies ever assembled. We sought to
integrate the findings with research on normal IMI and compare musical obsessions with
other obsessive-compulsive phenomena. The following are our main conclusions:
1. There are several reasons for regarding musical obsessions as true obsessions:
(a) forms of IMI can be identified that meet all DSM-5 diagnostic criteria for obsessions;

Page 29 of 60

29
(b) musical obsessions lead to avoidance and compulsions in the same way that occurs
for prototypic obsessions; and (c) musical obsessions are commonly comorbid with other
types of OC symptoms, as is typically the case for OCD in general (APA, 2013).

ip
t

2. The previous literature has been limited mainly to a scattering of isolated case
reports, which has been taken as suggesting that musical obsessions are rare. However,

cr

we were able to readily identify a large number of unpublished cases, obtained from

us

clinicians with expertise with OCD. It may be that musical obsessions are more common
than previously recognized.

an

3. Musical obsessions may be under-diagnosed and under-treated because

contemporary assessment methods do not sufficiently probe for such phenomena, and

clinicians may be unfamiliar with musical obsessions. Musical obsessions are sometimes

misdiagnosed as psychotic phenomena, which has led to inappropriate treatment (i.e.,

treatment.

te

antipsychotic medication). An accurate diagnosis is important for planning appropriate

Ac
ce
p

4. Musical obsessions differ from visual obsessions in that the content of the

obsession (i.e., the intrusive music) is not inherently aversive and does not violate the
persons system of beliefs or values. The content of visual obsessions is typically either
inherently aversive (e.g., horrific images of mutilated corpses), or violates the persons
values (e.g., blasphemous images). This raises the question of whether musical
obsessions and visual obsessions differ, at least to some extent, in their etiological
mechanisms. This conjecture, which awaits further investigation, is consistent with
research showing that OCD is etiologically heterogeneous (Taylor, 2011). That is, the
various symptoms of OCD (e.g., washing, checking, ordering) are shaped distinctive

Page 30 of 60

30
etiologic factors (environment and genetic) but also are shaped by factors that influence
all OC symptoms (Taylor, 2011).
5. IMI is common and yet musical obsessions are comparatively rare. Building on

ip
t

previous theorizing on OCD (e.g., Frost & Steketee, 2002), we propose a set of testable
hypotheses about the etiology of musical obsessions. Factors that influence the

cr

persistence or frequency of IMI may play a role but may be insufficient because

us

persistent, frequent IMI is not always distressing. Consistent with contemporary cognitive
models of OCD (Frost & Steketee, 2002), we propose that the persons beliefs and

an

appraisals of IMI play an important role in the etiology and maintenance of musical
obsessions. If a person appraises the IMI as threatening (e.g., as an indication of a life

threatening neurological problem), then he or she will become preoccupied and distressed

with the musical imagery. This may lead to functional impairment (e.g., failure to go to

te

the workplace because ringtones might be heard, or failure to fulfill household

responsibilities because of avoidance of supermarkets, in which background music is

Ac
ce
p

often played).

6. Although musical obsessions appear to respond to empirically supported

treatments for prototypic OC symptoms (e.g., ERP, SSRIs), there is evidence to suggest
that musical obsessions can be successfully treated with distraction-based interventions,
which are not among the recommended treatment of non-musical obsessions, such as
doubting/rumination and visual obsessions (Abramowitz et al., 2011).
7.2. Strengths and Limitations
There are various strengths and limitations to our review. In terms of strengths,
the present article is, to our knowledge, the first comprehensive review of musical

Page 31 of 60

31
obsessions, based on the largest set of case descriptions ever assembled. Our review is
comprehensive in that we sought to cover all the relevant literature, including, for
comparison with musical obsessions, the literatures on IMI in general and on visual

ip
t

obsessional imagery. In terms of weaknesses, the cases in our review were almost

entirely clinical (treatment-seeking) individuals, and so it is unclear whether the findings

cr

generalize to people who suffer from musical obsessions but do not seek treatment. A

us

further limitation is that there have been few studies comparing patients with musical
obsessions to other patient groups, and treatments have not been evaluated by means of

an

randomized controlled trials. A further limitation is that there have been few large-scale
empirical studies of IMI, and no empirical studies of the etiology of musical obsessions.

These limitations are unavoidable, given the lack of empirical attention to the neglected

clinical phenomena of musical obsessions. Despite these limitations, our review presents

te

the best available clinical description of musical obsessions, along with testable
hypotheses about etiology and treatment.

Ac
ce
p

7.3. Future Directions

Our review suggests several avenues for future investigation. Contemporary

measures of OC symptoms, including the current gold standard for assessing OC

symptomsthe Y-BOCSdo not adequately assess musical obsessions. This may have
led to an underestimation of the prevalence of the phenomena. Further research is needed
to develop psychometrically sound measures of musical obsessions and related
symptoms, such as music-related avoidance and compulsions, and beliefs and appraisals
of intrusive, uncontrollable musical imagery. The development of such measures will
facilitate epidemiological research on the lifetime prevalence and correlates of musical

Page 32 of 60

32
obsessions. Further research is also needed to investigate the cognitive-behavioral and
neurobiological bases of normal IMI in general, and musical obsessions in particular.
Research is also needed to investigate the promising treatments identified in the case

ip
t

studies, including the distraction-based interventions, which are quite different from

exposure-based empirically supported treatments for other types of obsessive-compulsive

Ac
ce
p

te

an

us

cr

phenomena.

Page 33 of 60

33
References
Abramowitz, J. S., Deacon, B. J., & Whiteside, S. P. H. (2011). Exposure therapy for
anxiety: Principles and practice. New York: Guilford.

ip
t

Ahuja, N. (2001). Two cases of musical obsessions. Proceedings of the 53rd Annual
National Conference of the Indian Psychiatric Society. Pune, India.

cr

Akhtar, S., Wig, N. N., Varma, V. K., Pershad, I., & Verma, S. K. (1975). A

British Journal of Psychiatry, 127, 342-348.

us

phenomenological analysis of symptoms in obsessive-compulsive neurosis.

an

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental

disorders (5th ed.). Arlington, VA: Author.

of Psychiatry, 39, 178-180.

Andrade, C., & Rao, N. S. K. (1997). Musical obsessions: A case report. Indian Journal

te

Bailes, F. A. (2007). The prevalence and nature of imagined music in the everyday lives
of music students. Psychology of Music, 35, 555-570.

Ac
ce
p

Beaman, C. P., & Williams, T. I. (2010). Earworms (stuck song syndrome): Towards a
natural history of intrusive thoughts. British Journal of Psychology, 101, 637-653.

Bennett, S. (2002). Musical imagery repetition (MIR). Masters thesis, Cambridge

University.

Berg, I. A. (1953). Observations concerning obsessive tunes in normal persons under

stress. Journal of Clinical Psychology, 9, 300-302.
Berrios, G. E. (1990). Musical hallucinations: A historical and clinical study. British
Journal of Psychiatry, 156, 188-194.
Cameron, O. G., & Wasielewski, P. (1990). Clomipramine treatment of possible atypical

Page 34 of 60

34
obsessive-compulsive disorder. Journal of Clinical Psychopharmacology, 10,
375.
Chauhan, N., Shah, R., & Grover, S. (2010). Obsessive auditory imagery: A case report.

ip
t

African Journal of Psychiatry, 13, 313.

de Silva, P. (1986). Obsessional-compulsive imagery. Behaviour Research and Therapy,

cr

24, 333-350.

us

Di Dio, A. S., Fields, M. C., & Rowan, A. J. (2007). Palinacousis - Auditory

perseveration: Two cases and a review of the literature. Epilepsia, 48, 1801-1806.

an

Dymond, S., & Roche, B. (2013). Advances in relational frame theory. Oakland, CA:
Context Press.

Ebbinghaus, H. (1885). ber das Gedchtnis: Untersuchungen zur experimentellen

Psychologie. Leipzig: von Duncker & Humblot.

te

El-Mallakh, R. S., & Walker, K. L. (2010). Hallucinations, pseudohallucinations, and

parahallucinations. Psychiatry, 73, 34-42.

Ac
ce
p

Emmelkamp, P. M. G., & Kwee, K. G. (1977). Obsessional ruminations: A comparison

between thought-stopping and prolonged exposure in imagination. Behaviour
Research and Therapy, 15, 441-444.

Evers, S., & Ellger, T. (2004). The clinical spectrum of musical hallucinations. Journal of
the Neurological Sciences, 227, 55-65.

Floridou, G. A., Williamson, V. J., & Mllensiefen, D. (2012). Contracting earworms:

The roles of personality and musicality. In E. Cambouropoulos, C. Tsougras, P.
Mavromatis, & K. Pastiadis (Eds.), Proceedings of the 12th international
conference of music perception and cognition (pp. 302-309). Thessaloniki,

Page 35 of 60

35
Greece.
Foa, E. B., Huppert, J. D., Leiberg, S., Langner, R., Kichic, R., Hajcak, G.,
Salkovskis, P. M. (2002). The obsessive-compulsive inventory: Development and

ip
t

validation of a short version. Psychological Assessment, 14, 485-496.

Fontenelle, L. F. (2008). Pareidolias in obsessive-compulsive disorder: Neglected

cr

symptoms that may respond to serotonin reuptake inhibitors. Neurocase, 14, 414-

us

418.

Frost, R. O., & Steketee, G. (Eds.). (2002). Cognitive approaches to obsessions and

an

compulsions: Theory, assessment and treatment. Oxford: Elsevier.

Gomibuchi, T., Gomibuchi, K., Akiyama, T., Tsuda, H., & Hayakawa, T. (2000).

Obsession of hearing music: From the viewpoint of Morita theory. Psychiatry and

Clinical Neurosciences, 54, 203-206.

te

Goodman, W. K., Rasmussen, S. A., Price, L. H., Mazure, C., Heninger, G. R., &
Charney, D. S. (1989). Manual for the Yale-Brown obsessive compulsive scale

Ac
ce
p

(rev.). New Haven, CT: Connecticut Mental Health Center.

Greenberg, D. (1987). Compulsive hoarding. American Journal of Psychotherapy, 41,

409-416.

Halpern, A. R., & Bartlett, J. C. (2011). The persistence of musical memories: A

descriptive study of earworms. Music Perception, 28, 425-431.

Hemming, J., & Altenmller, E. (2012). When an everyday-phenomenon becomes

clinical: The case of long-term earworms. In E. Cambouropoulos, C. Tsougras,
P. Mavromatis, & K. Pastiadis (Eds.), Proceedings of the 12th international
conference on music perception and cognition (pp. 419-420). Thessaloniki,

Page 36 of 60

36
Greece.
Hermesh, H., Konas, S., Shiloh, R., Reuven, D., Marom, S., Weizman, A., & GrossIsseroff, R. (2004). Musical hallucinations: Prevalence in psychotic and

ip
t

nonpsychotic outpatients. Journal of Clinical Psychiatry, 65, 191-197.

Hubbard, T. L. (2010). Auditory imagery: Empirical findings. Psychological Bulletin,

cr

136, 302-329.

us

Hyman, I. E., Burland, N. K., Duskin, H. M., Cook, M. C., Roy, C. M., McGrath, J. C., ...
Roundhill, R. F. (2013). Going Gaga: Investigating, creating, and manipulating

an

the song stuck in my head. Applied Cognitive Psychology, 27, 204-215.

Jacobs, L., Feldman, M., Diamond, S. P., & Bender, M. B. (1973). Palinacousis:

Persistent or recurring auditory sensations. Cortex, 9, 275-287.

Janata P. (2009). The neural architecture of music-evoked autobiographical memory

te

Cerebral Cortex, 19, 2579-2594.

Janet, P. (1903). Les obsessions et la psychasthnie. Paris: Alcan.

Ac
ce
p

Jaspers, K. (1913). Allgemeine Psychopathologie gie: Ein Leitfaden fr Studierende,

rzte und Psychologen. Berlin: Erstausgabe

Jenike, M. A. (1989). Obsessive-compulsive and related disorders: A hidden epidemic.

New England Journal of Medicine, 321, 539-541.

Knupp, S., de Oliveira, F., & Pelajo, C. F. (2010). Pediatric autoimmune neuropsychiatric
disorders associated with streptococcal infection (PANDAS): A controversial
diagnosis. Current Infectious Disease Reports, 12, 103-109.
Kraepelin, E. (1915). Psychiatrie: Ein Lehrbuch fr Studierende und rzte. Leipzig:
Verlag Von Johann Ambrosius Barth.

Page 37 of 60

37
Levitin, D. J. (2006). This is your brain on music. New York: Dutton.
Liikkanen, L. A. (2012a). Inducing involuntary musical imagery: An experimental study.
Musicae Scientiae, 16, 236-256.

ip
t

Liikkanen, L. A. (2012b). Musical activities predispose to involuntary musical imagery.

Psychology of Music, 40, 236-256.

cr

Liikkanen, L. A. (2012c). New directions for understanding involuntary musical imagery.

us

In E. Cambouropoulos, C. Tsougras, P. Mavromatis, & K. Pastiadis (Eds.),

Proceedings of the 12th international conference on music perception and

an

cognition (pp. 587-592). Thessaloniki, Greece.

Lipton, M. G., Brewin, C. R., Linke, S., & Halperin, J. (2010). Distinguishing features of

intrusive images in obsessive-compulsive disorder. Journal of Anxiety Disorders,

24, 816-822.

te

Marani, V. B., Aukst-Margeti, B., Grgi, N., & Kumi, E. (2011). Misdiagnosis and
exacerbation of unusual obsessive-compulsive disorder presentation with

Ac
ce
p

Risperidone and Clozapine in an adolescent girl a case report. Collegium

Antropologicum, 35 (Suppl. 1), 293-296.

Martinez, M., Marangell, L. B., & Martinez, J. M. (2008). Psychopharmacology. In R. E.

Hales, S. C. Yudofsky, & G. O. Gabbard (Eds.), Textbook of psychiatry (pp.
1053-1131). Washington, DC: American Psychiatric Association.

Matsui, T., Matsunaga, H., Ohya, K., Iwasaki, Y., Koshimune, K., Miyata, A., &
Kiriike, N. (2003). Clinical features in two cases with musical obsessions who
successfully responded to clomipramine. Psychiatry and Clinical Neurosciences,
57, 47-51.

Page 38 of 60

38
Mendhekar, D. N., & Andrade, C. (2009). Musical obsession: Repeated auditory imagery
of cell phone ring tone. Psychiatry and Clinical Neurosciences, 63, 591-592.
Miguel, E. C., Leckman, J. F., Rauch, S., do Rosario-Campos, M. C., Hounie, A. G.,

ip
t

Mercadante, M. T., Pauls, D. L. (2005). Obsessive-compulsive disorder

phenotypes: Implications for genetic studies. Molecular Psychiatry, 10, 258-275.

cr

Mohamed, W., Ahuja, N., & Shah, A. (2012). Palinacousis Evidence to suggest a post-

us

ictal phenomenon. Journal of the Neurological Sciences, 317, 6-12.

Mller, A. R. (2011). Misophonia, phonophobia, and exploding head syndrome. In A.

an

R. Mller (Ed.), Textbook of tinnitus (pp. 25-27). New York: Springer Science +
Business.

Mueller, A., Mitchell, J. E., Crosby, R. D., Glaesmer, H., & de Zwaan, M. (2009). The

prevalence of compulsive hoarding and its association with compulsive buying in

709.

te

a German population-based sample. Behaviour Research and Therapy, 47, 705-

Ac
ce
p

Murphy, T. K., Kurlan, R., & Leckman, J. (2010). The immunobiology of Tourettes
disorder, pediatric autoimmune neuropsychiatric disorders associated with
streptococcus, and related disorders: A way forward. Journal of Child and
Adolescent Psychopharmacology, 20, 317-331.

Nath, N., Bhattacharya, A., Hazarika, S., Roy, D., & Praharaj, S. K. (2013). A case of
early-onset obsessive-compulsive disorder with musical obsessions and comorbid
tic disorder. Journal of Neuropsychiatry and Clinical Neurosciences, 25, E14E15.
Obsessive Compulsive Cognitions Working Group. (2005). Psychometric validation of

Page 39 of 60

39
the obsessive belief questionnaire and interpretation of intrusions inventory Part
2: Factor analyses and testing of a brief version. Behaviour Research and
Therapy, 43, 1527-1542.

psychopathology (4th ed). Philadelphia, PA: Elsevier Saunders.

ip
t

Oyebode, F. (2001). Sims symptoms in the mind: An introduction to descriptive

cr

Pfizer, N., & Andrade, C. (1999). Isolated musical obsessions. Indian Journal of

us

Psychiatry, 41, 77-78.

Podoll, K. (2010). Musical palinacousis as an aura symptom in persistent aura without

an

infarction. In F. C. Rose (Ed.), Neurology of music (pp. 221-237). London:

Imperial College Press.

Praharaj, S. K., Goyal, N., Sarkar, S., Bagati, D., Sinha, P., & Sinha, V. K. (2009).

Musical obsession or pseudohallucination: Electrophysiological standpoint.

te

Psychiatry and Clinical Neurosciences, 63, 230-234.

Rapoport, J. L. (1980). The boy who couldnt stop washing: The experience and

Ac
ce
p

treatment of obsessive-compulsive disorder. New York: Dutton.

Reik, T. (1953). Haunting melody: Psychoanalytic experience in life and music. New
York: Farrar, Straus, and Young.

Sacks, O. (2007). Musicophilia. New York: Knof.

Saha, A. (2012). Musical obsessions. Industrial Psychiatry Journal, 21, 64-65.
Salkovskis, P. M. (1996). Cognitive-behavioral approaches to the understanding of
obsessional problems. In R. M. Rapee (Ed.), Current controversies in the anxiety
disorders (pp. 103-134). New York: Guilford.
Sanati, A. (2012). Pseudohallucinations: A critical review. Dialogues in Philosophy,

Page 40 of 60

40
Mental Health, and Neurosciences, 5, 42-47.
Sedman, G. (1966). A comparative study of pseudohallucinations, imagery and true
hallucinations. British Journal of Psychiatry, 112, 9-17.

ip
t

Speckens, A. E. M., Hackmann, A., Ehlers, A., & Cuthbert, B. (2007). Intrusive images
and memories of earlier adverse events in patients with obsessive compulsive

cr

disorder. Journal of Behavior Therapy and Experimental Psychiatry, 38, 411-422.

us

Stafford, T. (2012). Earworms: Why catchy tunes get trapped in our heads. Retrieved
July 16, 2013, from http://www.bbc.com/future/story/20120411-why-do-songs-

an

stick-in-our-heads/all

Taylor, S. (2011). Etiology of obsessions and compulsions: A meta-analysis and narrative

review of twin studies. Clinical Psychology Review, 31, 1361-1372.

Taylor, S., Abramowitz, J. S., McKay, D., & Cuttler, C. (2012). Cognitive approaches to

te

understanding obsessive-compulsive and related disorders. In G. Steketee (Ed.),

Oxford handbook of obsessive compulsive and spectrum disorders (pp. 233-250).

Ac
ce
p

New York: Oxford University Press.

Teismann, T., Michalak, J., Willutzki, U., & Schulte, D. (2012). Influence of rumination
and distraction on the therapeutic process in cognitive-behavioral therapy for
depression. Cognitive Therapy and Research, 36, 15-24.

Williamson, V. J., Jilka, S. R., Fry, J., Finkel, S., Mllensiefen, D., & Stewart, L. (2011).
How do earworms start? Classifying the everyday circumstances of involuntary
musical imagery. Psychology of Music, 40, 259-284.
Williamson, V. J., & Mllensiefen, D. (2012). Earworms from three angles: Situational
antecedents, personality predisposition and the quest for a musical formula. In E.

Page 41 of 60

41
Cambouropoulos, C. Tsougras, P. Mavromatis, & K. Pastiadis (Eds.),
Proceedings of the 12th international conference of music perception and
cognition (pp. 1124-1132). Thessaloniki, Greece.

ip
t

Zatorre, R. J., & Halpern, A. R. (2005). Mental concerts: Musical imagery and auditory
cortex. Neuron, 47, 9-12.

cr

Zungu-Dirwayi, N., Hugo, F., van Heerden, B. B., & Stein, D. J. (1999). Are musical

us

obsessions a temporal lobe phenomenon? Journal of Neuropsychiatry and

Ac
ce
p

te

an

Clinical Neurosciences, 11, 398-400.

Page 42 of 60

Appendix

cr

Appendix of Supplementary Materials for:

ip
t

us

Musical Obsessions: A Comprehensive Review of Neglected Clinical Phenomena

an

Correspondence: Steven Taylor, Ph.D., Department of Psychiatry, 2255 Wesbrook Mall,

University of British Columbia, Vancouver, BC, Canada, V6T 2A1. Email:

Ac
ce
p

te

taylor@mail.ubc.ca, Tel: 011-1-604-785-7558, Fax: 011-1-604-822-7756.

Page 43 of 60

ip
t
cr

Table A1. Previously published case studies of musical obsessions.

Age

Sex

Occupation

Musical
Obsession(s)
Tunes from Hindi
songs.

1. Ahuja (2001)

India

24

Trainee
physician

2. Ahuja (2001)

India

33

Homemaker

3. Akhtar et al.
(1975)
4. Andrade &
Rao (1997)

India

23

Student

India

30

Musician

5. Berg (1953):
Summary
of 10 cases

USA

6. Cameron &
Wasielewski
(1990)

USA

Duration of
Musical Obsession(s)
Two episodes; most
recent was 8 months

M
an

us

Country

Lyrics and tunes

from Hindi songs.
A tune popular at
the time.
Some musical
passages were more
troubling that others,
particularly Ravels
Bolero, a Chopin
nocturne, an operatic
piece by Verdi, and
some of the persons
own compositions.
Popular songs from
the 1920s-1950s.

18 months

Any music he had

recently heard.

2 years

Various popular
songs.

--

--

ce
pt
40%F

24

70%
employed,
30%
students
Student

Homemaker

Ac

7. Fontenelle
(2008)

Adult

Brazil

38

Notes, Including a Description of Other

Clinical Features (Current or Past)
No other OC symptoms. OCPD traits. Onset
while studying for exams.
Washing, sexual obsessions. OC personality
traits. Marked impairment in daily life.
--

Over 5 years

Recognized the musical imagery as his own

thoughts but regarded them as abnormal. Tried
to resist with little success, although
sometimes he was successful by concentrating
on other melodies. Also reported intrusive
verses of poetry, and checking and ordering
compulsions. OCPD traits. No evidence of
neurological problems.

--

All patients were in counseling for various

stress-related problems unrelated to musical
obsessions (e.g., distress regarding financial
problems).
No other psychopathology. Musical
obsessions started 1 month after he received a
blow to the head, but a comprehensive
evaluation, including EEG, revealed no
neurological abnormalities.
Distressing pareidolias (illusory images seen
out of shapes; e.g., seeing gorillas in the floor
tiles). Auditory and visual images of a dead
friend (auditory imagery of his name and
visual imagery of the imagined funeral).
Aggressive obsessions, washing, ordering, and
hoarding rituals. Depression,
depersonalization and derealization. Normal

ed

Case

Page 44 of 60

ip
t
19

Student

9. Gomibuchi et
al. (2000)
Case 2

Japan

18

Student

Croatia

17

11. Matsui et al.

(2003)
Case 1

Several
months

Japan

20

M
an

ed

Music he had heard

to the previous day.

Several
months

Student

Popular songs and

jingles.

One month
(approx.)

Student

Popular songs,
jingles, and other
well-known
melodies.

2 years

ce
pt

Ac

10. Marani et al.

(2011)

Various melodies he
had recently heard.

cr

Japan

us

8. Gomibuchi et
al. (2000)
Case 1

EEG.
Stress-related onset (studying for exams).
Tried to suppress the musical imagery, with
little success. Musical intrusions were most
frequent and distressing when he was trying
hard to concentrate while studying. OCPD
traits (highly perfectionistic). Misophonia
(low tolerance for sounds; minor sounds
evoked distress).
Stress-related onset (studying for exams).
Used to listen to music while studying because
it helped block out extraneous noises. Then
tried to study without listening to music but
began experiencing musical obsessions.
Musical intrusions were most frequent and
distressing when he was trying hard to
concentrate while studying. OCPD traits
(highly perfectionistic), mild checking
compulsions.
Grades at school dropped significantly due to
musical obsessions. She often attempted to
suppress the symptoms but was unable to do
so. Intense anxiety, insomnia,
depersonalization, derealization, along with
harming and sexual obsessions. Neurological
examination, including EEG, was normal.
Initially misdiagnosed as being acutely
psychotic.
Intrusions were particularly troublesome when
he was studying for exams. He described the
symptoms as a broken tape recorder
repeatedly producing the same songs and
music all day long. It was monotonous and
interfered with his normal routine, academic
functioning, and usual social activities or
relationships. Exacerbated by stress (studying
for exams). Coping responses to musical

Page 45 of 60

ip
t
cr
us
13. Mendhekar &
Andrade
(2009)

India

28

Retail sales

M
an

Japan

Current and old

popular songs.

2 years

ce
pt

ed

12. Matsui et al.

(2003)
Case 2

Student

Various cell phone

ring tones, occurring
6-8 hrs/day.

4 months

--

Recently heard

3 years

Ac

22

14. Nath et al.

India

22

obsessions consisted of covert compulsions

(suppression, neutralization). OCPD traits
(perfectionistic), anxious, and depressed
(dysthymic disorder). The more he worried
about, and tried to suppress, the musical
imagery, the more persistent it became.
Avoided classes and social activities due to
musical obsessions. Experienced
depersonalization. Initially misdiagnosed as
having schizophrenia. Neurological exam
including EEG was normal. ENT normal.
Premorbid tender and obedient character.
Onset after he quit his job due to interpersonal
conflict (stress-related). He often attempted to
suppress the symptoms by thinking about
another melody (i.e., suppress, neutralize),
although he usually was unable to do so.
Obsessions worsened when he started living
alone. Neurological exam normal, including
normal EEG. Comorbid major depressive
disorder.
Feared and avoided exposure to ring tones. He
wore earplugs, used distraction by thinking of
other tunes, kept his phone in silent mode, and
asked others to do the same, avoided places he
might hear the ringing of cell phones. Prior to
onset of musical obsession he had enjoyed
downloading and changing ring tones on his
phone. He would devote 2-3 hrs every 2-3
days to this activity (i.e., had been spending a
lot of time listening to ringtones). History of
counting, checking, and washing. At the time
of presentation he had mild comorbid
checking. Musical obsessions interfered with
his ability to study and became depressed as a
result.
Attempted to resist the musical imagery but

Page 46 of 60

ip
t
cr

tunes.

India

25

Homemaker

16. Praharaj et al.

(2009)

India

21

--

17. Rapoport
(1980)

USA

A single Tamil film

song running
through her head for
8 hrs/day

Arose suddenly 4 days

prior to seeking
treatment, along with a
sore throat and fever,
but did not abate when
the infection was
successfully treated
with erythromycin.
5 years

M
an

15. Pfizer &

Andrade
(1999)

us

(2013)

ce
pt

ed

Sections of music
from various
popular Hindi films,
lasting from 2-45
min per episode,
occurring 30-35
times/day, nearly
every day.
Occasionally a new
song he heard
replaced an older
one.
Popular music,
movie scores.

Student

Farmer

7 years

Ac

16

18. Rapoport
(1980)

USA

60

Six or more notes of

a fiddle tune (a dull
short jig).

31 years

was unsuccessful. Vocal tics, washing,

harming obsessions, ordering, hoarding,
sexual obsessional images. Initially
misdiagnosed as having schizophrenia. No
history of neurological problems.
Her attempts at controlling the intrusion were
unsuccessful. No past history of
psychopathology. Musical obsessions
persisted despite no evidence of a general
medical condition. Reported feeling
depersonalized because of the obsession.

Tunes commonly intruded when he was alone.

He was partially successful in temporarily
banishing them by distraction such as
engaging in games or conversations, but the
intrusions returned. Washing, checking,
doubting, intrusive visual obsessions, severe
depression. Normal physical examination.
Normal EEG.

Past history of checking, counting, washing,

and superstitious rituals (only even numbers
were acceptable). Was in therapy for 4 years
for anxiety and adjustment problems but never
disclosed his musical obsessions. Efforts at
suppressing imagery was unsuccessful.
Musical obsessions greatly interfered with
everyday functioning. Normal CT.
Kept the musical obsession secret from others.
His spouse observed that he looked
preoccupied while he was experiencing the
obsessions. No evidence of neurological or

Page 47 of 60

ip
t
cr
30

--

20. ZunguDirwayi et al.

(1999) Case
1

South
Africa

59

--

us

India

Music that he or
others had
composed,
associated with
intrusive images of
him humming the
tunes.
Various tunes.

4 years

M
an

19. Saha (2012)

audiologic problems, and no evidence of

psychosis. Experienced the tune as a constant
unpleasantness, bore, and irritation. Described
it as living my life through the noise.
Attempted to resist with little success.
Distraction with other melodies was
successful, at least in the short term. Intrusive
verses of poetry, along with checking,
ordering symptoms.

Ac

ce
pt

ed

20 years. Developed in
Tried to get rid of the musical obsessions by
the context of major
substituting them with other thoughts or tunes.
depression following
Normal EEG.
her husbands death 20
years previously.
Obsessions persisted
after depression abated.
South
29
F
-Unspecified tune.
Intrusive tune
Unconscious for 4 days due to head injury.
21. ZunguDirwayi et al.
Africa
developed 5 months
One month of retrograde amnesia. Post-injury
(1999) Case
after a closed head
personality changes (increased aggressiveness
2
injury. Intrusion had
and apathy) with little insight as to the
persisted for a month at
changes. CT scan was normal. Unsuccessful
the time of assessment.
attempts at thought suppression.
-- = Not reported. Gomibuchi et al. (2000) presented, for illustrative purposes, 2 of 5 of their cases of musical obsessions (4 M, 1 F, ages 18-22). References to all
cited articles appear in the reference list of the main article.

Page 48 of 60

ip
t
Sex

Occupation

Musical
Obsession(s)

1. BCR
Case 1

USA

21

Student

2. BCR
Case 2

USA

59

Unemployed
(former
cleric)

3. CA

India

23

Songs on his
iPod, which
were mostly
popular songs.
If someone
was to sing a
song in
passing, and
he knew the
song, it would
also become a
musical
obsession.
Numerous
types of songs
(e.g., jingles,
Christmas
songs,
hymnals,
popular music
on the radio).
Not reported.

4. DM
Case 1

USA

Duration of
Musical
Obsession(s)
--

ed

ce
pt
M

Unemployed

Student

A pop song
that was
heavily played
on the radio.

Notes, Including Details of Other Clinical

Features (Current or Past)

us

Age

M
an

Country

Ac

Case

cr

Table A2. Unpublished cases of musical obsessions from the authors clinical files.

Would hear a song involuntarily playing in his

head and then need to play it out or listen to it in
full before moving on otherwise it would be
stuck all day. Washing, hoarding, doubting,
obsessions (somatic, aggressive, sexual).
Anorexia nervosa, mood disorder NOS, GAD.

Intermittently
since teens.

Fear that a song will get stuck in his head and

then he will be unable to function. No other OC
symptoms. Recurrent, moderate major depressive
disorder.

Intermittently,
sometimes
lasting a day
or more.

Child-onset OCD and comorbid bipolar disorder,

for which the patient was in a depressive phase at
the time of assessment and treatment. OC
symptoms included musical and visual
obsessions. The latter consisted of sexual and
violent obsessions.
Avoided exposure to music for fear of developing
new intrusive song. Sexual and aggressive
obsessions, along with compulsions to confess
unwanted thoughts. Past history of separation
anxiety and mild attention difficulties.

2 months

Page 49 of 60

ip
t
cr

5. DM
Case 2
6. DM
Case 3

USA

37

Lawyer

USA

49

Lawyer

7. DS
Case 1
8. DS
Case 2
9. DS
Case 3
10. DS
Case 4

Canada

25

Student

Canada

67

Canada

43

Retired
businessman
Homemaker

Canada

83

Retired
lawyer

11. DS
Case 5
12. ECM&
ADM
Case 1
13. ECM&
ADM
Case 2
14. ECM&
ADM
Case 3
15. ECM&
ADM
Case 4
16. ECM&
ADM
Case 5
17. ECM&
ADM
Case 6

Canada

32

Retail sales

Brazil

22

Student

Brazil

25

Factory
worker

Not reported.

10 years

Washing, ordering, hoarding, and obsessions

(aggressive, sexual, religious).

Brazil

28

Office
supervisor

Not reported.

14 years

Washing, ordering, hoarding.

4 months

--

OCPD traits (perfectionistic). Symmetry,

ordering, and counting.
Checking and religious obsessions.

Lullabies.

--

Washing.

Funeral dirges.

3 months.
Musical
obsession
started after
the death of
his wife.
--

Depression, ordering.

Any song she

recently heard.
Not reported.

Washing, checking. Past history of abuse of a

prescribed anxiolytic.
Violent imagery and harming obsessions. Past
history of depression.

us

Intermittently
for 2 years

M
an

ed

ce
pt

Ac

Heavy metal
songs
Popular songs
and TV
themes.
Bach, various
piano music.
Choral music.

--

8 years

Musical obsessions exacerbated by stress. OCPD

traits (perfectionistic). Ordering rituals.
Washing, ordering, hoarding, and obsessions
(aggressive, sexual, religious).

Brazil

31

Office
assistant

Not reported.

16 years

Washing, ordering, hoarding, and obsessions

(aggressive, sexual, religious).

Brazil

21

Computer
industry

Not reported.

4 years

Hoarding, ordering, and obsessions (aggressive,

sexual, religious).

Brazil

31

Architect

Not reported.

6 years

Washing, ordering, and obsessions (aggressive,

sexual, religious).

Page 50 of 60

ip
t
cr

37

Business
consultant

Not reported.

19 years

Brazil

45

Researcher

Not reported.

20 years

Washing and ordering.

Brazil

34

Advertising

Not reported.

23 years

Washing, hoarding, ordering, and obsessions

(aggressive, sexual, religious).

Brazil

22

Telemarketer

Not reported.

1 year

Washing, ordering, and aggressive obsessions.

Brazil

60

Homemaker

Not reported.

53 years

Washing, ordering, and obsessions (aggressive,

sexual, religious).

Brazil

25

Lawyer

Not reported.

13 years

Washing, hoarding, ordering, and obsessions

(aggressive, sexual, religious).

Brazil

58

Retired

Not reported.

18 years

Washing, ordering, and obsessions (aggressive,

sexual, religious).

Brazil

42

Independently
wealthy*

Not reported.

24 years

Washing, hoarding, ordering, and obsessions

(sexual and religious)

Brazil

46

Office
worker

Not reported.

20 years

Washing, ordering, and aggressive obsessions.

Brazil

54

Homemaker

Not reported.

24 years

Washing, hoarding, ordering, and obsessions

(aggressive, sexual, and religious)

Brazil

43

Homemaker

Not reported.

23 years

Washing, ordering, and obsessions (aggressive,

sexual, and religious)

Brazil

32

Secretary

Not reported.

20 years

Washing, hoarding, ordering, and aggressive

obsessions.

Brazil

60

Office

Not reported.

45 years

Washing, hoarding, ordering, and obsessions

M
an

ed

ce
pt

Washing, ordering, and obsessions (aggressive,

sexual, religious).

us

Brazil

Ac

18. ECM&
ADM
Case 7
19. ECM&
ADM
Case 8
20. ECM&
ADM
Case 9
21. ECM&
ADM
Case 10
22. ECM&
ADM
Case 11
23. ECM&
ADM
Case 12
24. ECM&
ADM
Case 13
25. ECM&
ADM
Case 14
26. ECM&
ADM
Case 15
27. ECM&
ADM
Case 16
28. ECM&
ADM
Case 17
29. ECM&
ADM
Case 18
30. ECM&

Page 51 of 60

36.

37.

38.

39.

40.

41.

cr

ip
t
35.

Homemaker

Not reported.

33 years

Washing, hoarding, ordering, and obsessions

(aggressive, sexual, and religious)

Brazil

27

Marketing

Not reported.

22 years

Brazil

25

Lawyer

Not reported.

14 years

Washing, hoarding, symmetry

Brazil

60

Retired

Not reported.

10 years

Symmetry

Brazil

33

Lawyer

Not reported.

26 years

Washing, hoarding, symmetry, and obsessions

(aggressive, sexual, and religious)

Brazil

22

Student

Not reported.

3 years

Washing, symmetry, and obsessions (aggressive,

sexual, and religious)

Brazil

25

Admin.

Not reported.

10 years

Washing, symmetry

Brazil

68

Admin.

Not reported.

53 years

Washing, symmetry, and obsessions (aggressive,

sexual, and religious)

Brazil

18

Student

Not reported.

10 years

Washing, symmetry

us

M
an

34.

53

ed

33.

(aggressive, sexual, and religious)

Brazil

ce
pt

32.

worker

Ac

31.

ADM
Case 19
ECM&
ADM
Case 20
ECM&
ADM
Case 21
ECM&
ADM
Case 22
ECM&
ADM
Case 23
ECM&
ADM
Case 24
ECM&
ADM
Case 25
ECM&
ADM
Case 26
ECM&
ADM
Case 27
ECM&
ADM
Case 28
ECM&
ADM
Case 29
ECM&
ADM
Case 30

10

Symmetry

Brazil

26

Admin.

Not reported.

6 years

Symmetry

Brazil

29

Hair stylist

Not reported.

14 years

Washing, hoarding, symmetry, and obsessions

(aggressive, sexual, and religious)

Page 52 of 60

ip
t
cr

48

Teacher

Not reported.

1 year

Brazil

21

Student

Not reported.

6 years

Symmetry, and obsessions (sexual and religious)

Brazil

42

Diplomat

Not reported.

25 years

Symmetry and violent obsessions

Brazil

27

Communications

Not reported.

17 years

Washing, symmetry

Brazil

20

Teacher

Not reported.

8 years

Washing, hoarding, symmetry, and obsessions

(aggressive, sexual, and religious)

Brazil

38

Homemaker

Not reported.

23 years

Washing

Brazil

36

Dentist

Not reported.

22 years

Washing, hoarding, symmetry, and obsessions

(aggressive, sexual, and religious)

Brazil

27

Admin.

Not reported.

17 years

Symmetry, and obsessions (aggressive, sexual,

and religious)

Brazil

33

Railway
worker

Not reported.

2 years

Washing, hoarding, symmetry, and obsessions

(aggressive, sexual, and religious)

Brazil

30

Saleswoman

Not reported.

1 year

Washing, hoarding, symmetry

Brazil

29

Technical
support

Not reported.

7 years

Washing, hoarding, symmetry, and obsessions

(aggressive, sexual, and religious)

Brazil

24

Actress

Not reported.

9 years

Symmetry and violent obsessions

Republic

24

Student

Korean pop

4 years

Musical obsessions were accompanied by mild

M
an

ed

ce
pt

Symmetry, and obsessions (aggressive, sexual,

and religious)

us

Brazil

Ac

42. ECM&
ADM
Case 31
43. ECM&
ADM
Case 32
44. ECM&
ADM
Case 33
45. ECM&
ADM
Case 34
46. ECM&
ADM
Case 35
47. ECM&
ADM
Case 36
48. ECM&
ADM
Case 37
49. ECM&
ADM
Case 38
50. ECM&
ADM
Case 39
51. ECM&
ADM
Case 40
52. ECM&
ADM
Case 41
53. ECM&
ADM
Case 42
54. JSK&

11

Page 53 of 60

ip
t
cr

MJH
Case 1

of Korea

55. JSK&
MJH
Case 2

Republic
of Korea

21

Student

Korean pop
song.

56. JSK&
MJH
Case 3

Republic
of Korea

30

Unemployed

Korean pop
song.

9 months

57. JSK&
MJH
Case 4

Republic
of Korea

20

Student

Korean pop
song.

4 years

58. JSK&
MJH
Case 5

Republic
of Korea

18

Student

Korean pop
song.

6 years

59. JSK&
MJH
Case 6

Republic
of Korea

31

Office
worker

Korean pop
song.

9 years

M
an

3 years

ed
ce
pt

depression. Symptoms were aggravated while

studying for exams. Patient also had aggressive,
sexual obsessions and related checking as
concurrent symptoms.
Musical obsessions started after conflicts with
friends at school. Patient also had aggressive
obsessions without compulsion, somatic
obsessions, pathological doubt and idea of
reference as concurrent symptoms.
After listening to music, the melody of music
continued for an hour. Due to the somatic
obsessions and related idea of reference he had
cosmetic surgery on his nose. Comorbid major
depressive disorder and social anxiety disorder
When he experienced musical obsessions, he also
had to perform his activities according to the
rhythm of them. Patient also had repeating,
checking, contamination, cleaning, perfectionism
and impulsivity as concurrent symptoms.
Comorbid bipolar disorder NOS.
Korean pops which he usually listens to replayed
in his head for several hours. These musical
obsessions did not disturb his concentration or
activities of daily life. Patient also had
pathological doubt, checking, aggressive and
sexual obsessions as concurrent symptoms.
Musical obsessions started after job-related stress.
The melody of the music that she heard in the
morning replayed all day long. She began bingeeating to avoid the stress due to the musical
obsessions. Recently musical obsessions were
aggravated by the conflict with a colleague at
work and she couldnt concentrate on her work.
Patient also had checking and hoarding
compulsions. Comorbid bulimia nervosa, alcohol
abuse and depression.

us

song.

Ac

12

Page 54 of 60

ip
t
Unemployed

61. KO

Canada

32

Student &
teacher

62. LF

Brazil

29

Student

63. LH

USA

cr

Sad songs
replaying in
his head. He
feared that
they would
provoke bad
events in
reality
Parts of
various songs
intrude
randomly

--

Any recently
heard song
with chorus

10 years

Any recently
heard song.

Intermittent
episodes since
adolescence.

--

ce
pt
Ac
24

Admin.
assistant and
graduate
student.

Duration of up to 12 hrs/day. Onset frequently

associated with sad memories. Sometimes the
intrusive musical imagery was so intense that he
difficulty following conversations. He consulted a
psychiatrist for fears that he might have
schizophrenia. Comorbid symmetry and
superstitious rituals, and body focused repetitive
behaviors (skin picking and nail biting).
Songs can repeat in a loop up to 4hrs per day
every day, interferes with work and writing,
incapable of concentrating. No particular trigger
but only occur when stressed.Comorbid harming
obsessions.
History of obsessions with aggressive content at
age 9, which disappeared soon afterwards. At age
14, she had her first panic attack and developed,
since then, chronic panic disorder with
agoraphobia. OCD, including sexual/religious and
aggressive obsessions reappeared acutely at age
19, together with the onset of musical obsessions.
For quite awhile, musical obsessions remained
her most significant symptom. They were worse
when she was either relaxed or very stressed. Of
note, she did not endorse any type of overt or
covert compulsions. At her worse, her Y-BOCS
obsessions score was 14 and BDI 26. She had her
first depressive episode at age 19, 2 months
before the onset of OCD. In addition, she
developed chronic and persistent
depersonalization disorder (not ascribed to OCD
or panic attacks) since age 23.
Musical obsession was a negative side effect of a
coping strategy of singing to herself to distract
from other obsessions. Began hearing songs
whenever not engaged in focused conversation.
Musical obsession caused significant interference,
especially at night, contributing to insomnia.

us

33

M
an

Canada

ed

60. KO

13

Page 55 of 60

ip
t
cr

14

Ac

ce
pt

ed

M
an

us

Visual obsessions (violent and horrific images),

washing, checking, hoarding, specific phobias of
spiders and blood. Past history of anorexia
nervosa and bulimia nervosa.
Initials of cases refer to the authors of the present article. All cases were assessed with the Y-BOCS. *Classified as employed because the person has an income.

Page 56 of 60

ip
t
cr

15

Table A3. Summary of treatments for previously published case studies of musical obsessions.

6. Gomibuchi et al.
(2000) Case 1
7. Gomibuchi et al.
(2000) Case 2

us

Clomipramine up to 100 mg/day for 12 weeks. OCD was rated as very much improved according to both the CGI (score=1) and
YBOCS (score=5). Given the very low scores it is most likely that her musical obsession abated with treatment although this is not
explicitly stated by the author.
Diazepam (6mg/day) along with advice not to study too hard and not to attempt to suppress musical imagery. Five sessions in all.
Reported being symptom free at 5-year followup.
Bromazepam (6mg/day) along with advice not to study too hard and not to attempt to suppress musical imagery. Three sessions in all.
Musical obsession diminished within a week. Relapsed at 5-year followup but then improved after he re-implemented the strategy of
not trying to suppress musical imagery.
Failed to respond to, or worsened, on antipsychotics (risperidone, clozapine, ziprasidone, haloperidol, flufenazine). Musical obsessions
remitted entirely after 2 months of fluvoxamine (200 mg/day). Dose increased to 300 mg/day. Symptom free at 6-month followup.
No response to haloperidol or risperidone. Responded to clomipramine, up to 150 mg/day for 6 weeks. Rated as much improved on the
CGI.
No response to haloperidol (6 mg/day for 2 months). Then received 8 weeks of clomipramine (up to 150 mg/day), after which he was
rated as much improved on the CGI.
Treated with a combination of fluvoxamine (200 mg/day) and clomipramine (75 mg/day), after which the patient reported a 90%
improvement. No residual functional impairment at 3-month followup.
Symptoms worsened when treated with olanzapine (10 mg/day). Was then treated with clomipramine (up to 150 mg/day) augmented
with flupenthixol (1 mg/day). This markedly improved tics and all OC symptoms except for musical obsessions, for which treatment
had no effect. Patient reported a 20% improvement in musical obsessions after treatment with a combination of thought-stopping, ERP,
and Morita therapy.
Clomipramine (75 mg/day) and alprazolam (.75 mg/day), along with antibiotic and antipyretic medication for sore throat and fever.
The sore throat and fever remitted within a week but the musical obsessions remained unchanged. Discontinued psychotropic
medications due to side effects. Dropped out of treatment. The patient wanted to try to control the obsessions on her own.
Musical obsession was resistant to pharmacotherapy (fluvoxamine 300mg/day augmented with respiridone 2mg/day). However, the
other OC symptoms and depression responded to the fluvoxamine/respiridone combination. There was some reduction in frequency
and duration of musical obsessions by applying thought-stopping. After 6 weeks of practicing thought-stopping, musical obsessions
reduced from up to 45 min per episode to no more than 30 min per episode. Frequency reduced from 30-35 times/day to 20-25

Ac

8. Marani et al.
(2011)
9. Matsui et al.
(2003) Case 1
10. Matsui et al.
(2003) Case 2
11. Mendhekar &
Andrade (2009)
12. Nath et al. (2013)

M
an

5.

Clomipramine up to 250 mg/day for 14 weeks. Statistical and clinically significant improvement. Patient was able to return to school.
Prior to receiving clomipramine the patient had failed to respond to other, unspecified medications.

ed

4.

Treatment and Outcome

Distraction and fluoxetine (40 mg/day).
Fluvoxamine (200 mg/day), buspirone (40 mg/day).
Patient was offered pharmacotherapy but declined. Preferred to try to control the obsessions on his own.

ce
pt

1.
2.
3.

Case
Ahuja (2001)
Ahuja (2001)
Andrade & Rao
(1997)
Cameron &
Wasielewski
(1990)
Fontenelle
(2008)

13. Pfizer & Andrade

(1999)
14. Praharaj et al.
(2009)

Page 57 of 60

ip
t
cr

Ac

ce
pt

ed

17. Zungu-Dirwayi
et al. (1999)
Case 1
18. Zungu-Dirwayi
et al. (1999)
Case 2

us

16. Saha (2012)

times/day.
Clomipramine plus L-tryptophan. Despite a 31 year history of musical obsessions, the patient was symptom free after 3 months of
treatment.
Combined fluvoxamine (200 mg/day) and fluoxetine (60 mg/day), augmented with lithium (900 mg/day). Also received
psychotherapy, ERP, and systematic desensitization. Clinically significant response after 8 weeks of treatment and was almost
completely remitted at 6 month followup.
Failed to respond to trials of various medications including SRIs (fluoxetine, paroxetine, citalopram, and clomipramine). She also
failed to respond to augmentation of SRIs with risperidone or gabapentin, and was unable to tolerate therapeutic doses of
carbamazepine or valproic acid.
The patient was offered pharmacotherapy but declined.

M
an

15. Rapoport (1980)

16

Page 58 of 60

ip
t
cr

Table A4. Treatments for the authors cases of musical obsessions.

4. DM
Case 1
5. DM
Case 2
6. DM
Case 3

7. DS
Case 1
8. DS
Case 2
9. DS
Case 3
10. DS
Case 4
11. DS
Case 5
12. JSK

--

us
M
an

Psychoeducation about intrusive thoughts, ERP consisting of refraining from playing the song in his head to
completion. That in turn, evolved into a compulsion; when a song began to play he felt compelled to stop playing the
song or interrupt it. Treatment then shifted toward acceptance of intrusive musical imagery. That is, he was encouraged
to accept that songs played in his head and to not try to resist the intrusive imagery.
Exposure to catchy songs and jingles, with continued activity (no avoidance of other tasks if/when song gets
stuck). Acceptance when song gets stuck.
Treatment focused on OCD and comorbid depression. When the patient presented to CA, a range of treatments tried.
OC symptoms improved with a combination of clomipramine (150 mg/day), clozapine (300 mg/day), and valproate
(1,500 mg/day). However, depressive symptoms persisted. Depression improved with a course of ECT. Patient also
appeared to respond to intranasal ketamine (0.5 mg/kg/day). Musical obsessions persisted in an attenuated form despite
improvements in other OCD and mood symptoms.
2 sessions of rapid competing music sing-alongs (with clinician) and singing target obsession song incorrectly

ed

2. BCR
Case 2
3. CA

Treatment and Outcome

ce
pt

1. BCR
Case 1

Pretreatment
global
severity
(CGI-S)
5

Outcome
(CGI-I)

3
--

2
1

4 sessions of rapid competing music sing-alongs (with clinician) and singing target obsession song incorrectly. Patient
also had treatment sessions for other OC symptoms. Responded rapidly to treatment, with reductions in all symptoms.
8 sessions of rapid competing music sing-alongs (with clinician) and singing and humming target obsession songs
incorrectly. Patient also had treatment sessions for other OC symptoms. Client struggled to fully conquer musical
obsessions and other obsessions.
6 months of weekly CBT (including ERP).

9 months of weekly CBT (including ERP).

12 months of weekly or twice-weekly CBT (including ERP). Symptom-free at 1-year followup.

1
3

8 months of weekly CT largely related to mourning the death of his wife, and also CT and ERP for ordering symptoms
at home that worsened after the death of his wife. Musical obsessions were not directly treated but they abated as a
result of the CT related to mourning.
6 months of weekly CBT (including ERP).

Escitalopram (60mg/day).

7
5

Ac

Case

17

Page 59 of 60

ip
t
cr

Quetiapine (100mg/day), fluoxetine (80mg/day).

Aripiprazole (30mg/day).

Aripiprazole (10mg/day), venlafaxine (75mg/day), lamotrigine (50mg/day), valproic acid (325mg/day). Medications
were administered simultaneously to treat OCD comorbid with bipolar disorder NOS.
Aripiprazole (10mg/day), sertraline (200mg/day).

Quetiapine(100mg/day), fluoxetine(80mg/day). CBT

us

M
an

Case 1
13. JSK
Case 2
14. JSK
Case 3
15. JSK
Case 4
16. JSK
Case 5
17. JSK
Case 6
18. KO
19. LF

18

CBT. Over the course of therapy, Y-BOCS scores dropped from 35 to 21.
-All OC symptoms including musical obsessions improved significantly with clomipramine, 150 mg/day (post-treatment Y-BOCS=5).
20. LH
6
ERP for main targets (exposure to and altering songs), then distraction and label as OCD and move on strategies for
1
any target aspects with SUDS ratings 3 or below.
21. NA
4
Described in Table A3. The ratings were obtained as supplementary unpublished information.
1
22. NA
5
Described in Table A3. The ratings were obtained as supplementary unpublished information.
1
Clinical Global Impression (CGI) scales: CGI-S (pretreatment severity for all pretreatment psychopathology): 1, normal, not at all ill; 2, borderline mentally ill;
3, mildly ill; 4, moderately ill; 5, markedly ill; 6, severely ill; or 7, extremely ill. CGI-I (treatment-related improvement for musical obsessions): 1, very much
improved; 2, much improved; 3, minimally improved; 4, no change; 5, minimally worse; 6, much worse; or 7, very much worse. -- = Not reported.

Ac

ce
pt

ed

6
--

Page 60 of 60