Journal of Music Therapy, 53(1), 2016, 1–26

doi:10.1093/jmt/thv017; Advance Access publication December 7, 2015

© the American Music Therapy Association 2015. All rights reserved.
For permissions, please e-mail: [email protected]

Music Therapy Assessment Tool for

Awareness in Disorders of Consciousness
(MATADOC): Reliability and Validity of a
Measure to Assess Awareness in Patients
with Disorders of Consciousness
Wendy L. Magee, PhD
Boyer College of Music and Dance, Temple University,
Philadelphia, PA, and Royal Hospital for Neuro-Disability, London

Richard J. Siegert, PhD

School of Rehabilitation and Occupation Studies and School of
Public Health and Psychosocial Studies, Auckland University of
Technology, Auckland, New Zealand

Steve M. Taylor, MSc (hons)

Department of Biostatistics and Epidemiology, Auckland University
of Technology, Auckland, New Zealand

Barbara A. Daveson, PhD

King’s College London, Cicely Saunders Institute, Department of
Palliative Care, Policy, and Rehabilitation

Gemma Lenton-Smith, MA Music Therapy

Ealing Music Therapy, London

Background:  Prolonged Disorders of Consciousness (PDOC) describes

a population where a consciousness disorder has persisted for at least
four weeks post injury but is still under investigation. Complex motor, sen-
sory, communication, and cognitive impairments cause challenges with
diagnosis, assessment, and intervention planning. Developing sensitive,
reliable, and valid measures is a central concern. The auditory modality is
the most sensitive for identifying awareness; however, the current stand-
ardized behavioral measures fail to provide adequate screening and meas-
urement of auditory responsiveness. The Music Therapy Assessment Tool

Address correspondence concerning this article to Wendy Magee, PhD, Temple

University Presser Hall, 2001 North 13th Street, Philadelphia, PA 19122. Phone:
215-204-1982. E-mail: [email protected]
2 Journal of Music Therapy

for Awareness in Disorders of Consciousness (MATADOC) is a recently

standardized measure for assessment with PDOC; however, psychomet-
ric values for two of its subscales require examination.
Objective:  To determine the measurement characteristics and proper-
ties of the MATADOC subscales two and three.
Methods:  In a convenience sample of 21participants with PDOC, a pro-
spective repeated measures study examined inter-rater reliability (IRR)
and test-retest reliability (TRR) for both subscales and internal consist-
ency of subscale two.
Results:  Overall, the items from the MATADOC subscales two and three
demonstrated good agreement across and within assessors, with some
variability on two identified items.
Conclusions:  The MATADOC is a standardized measure for assess-
ment of auditory responsiveness in PDOC. Psychometric limitations
for the two identified items may have resulted from variations in music
therapist clinical experience and training, leading to variations in the
administration and interpretation of PDOC patient responses to these
two MATADOC assessment items. Although its psychometric properties
could be improved, the MATADOC’s clinimetric properties make it a valu-
able assessment to guide clinical work for patients with PDOC.
Keywords:  music therapy; disorders of consciousness; brain injuries;
assessment; vegetative state; minimally conscious state

Introduction
Disorders of Consciousness (DOC) describes a continuum
of acquired conditions that stem from acquired profound brain
injury. Three primary conditions can be categorized under DOC.
Coma represents a state of unarousable unresponsiveness in which
there is no evidence of self or environmental awareness (Plum &
Posner, 1983). It is usually a temporary phase that progresses to
some level of consciousness. Vegetative State (VS) is character-
ized by spontaneous arousal and sleep-wake cycles but with no
evidence of awareness of self or environment, no interaction with
others, and no sustained, reproducible, purposeful, or voluntary
behavioral responses to sensory stimuli (Jennett & Plum, 1972).
Some nerve and spinal reflex responses are preserved, present-
ing as spontaneous, non-purposeful movements that are not goal
directed. Progression from VS leads to Minimally Conscious State
Vol. 53, No. 1 3

(MCS), where although consciousness remains severely dimin-

ished, the person demonstrates minimal but definite behavioral
evidence of self or the environment (Giacino et al., 2002). People
in MCS are able to demonstrate visual pursuit of moving objects,
orient away from noxious stimuli, and may be able to follow simple
one-step verbal commands and manipulate objects (Schnakers &
Majerus, 2012), although responses typically remain inconsistent.
The nomenclature of “VS” and “MCS” is the most widely accepted
at the current time, although there is a call to revise “Vegetative
State,” given the negative connotations in the English language with
the word “vegetable,” to the less emotive “Unresponsive Wakefulness
Syndrome” (Laureys et al., 2010). Several subcategories of MCS have
been proposed to discriminate between people who can demonstrate
“higher-level” responses (e.g., command following; communication
of “yes” or “no”) and those who show less sophisticated or a more
limited repertoire of behaviors (e.g., visual tracking and/or orienting
away from noxious stimuli only) (Bruno, Vanhaudenhuyse, Thibaut,
Moonen, & Laureys, 2011). Descriptors such as MCS+ and MCS–
offer sensitive differentiations within MCS and thus make sense to cli-
nicians working with such complex patients who witness subtle varia-
tions in patient presentation. However, the revised terminologies are
not yet recommended by authoritative sources. The term “Emerging”
(i.e., from DOC) describes individuals who show purposeful, repro-
ducible behaviors that are related to the external environment and
are more complex in nature. Most recently, DOC has been revised to
“Prolonged Disorders of Consciousness” (PDOC) to describe a DOC
that has persisted for at least four weeks post injury but is still under
investigation (Royal College of Physicians, 2013).
The incidence of PDOC is difficult to determine accurately; how-
ever, recent estimates stand at 46 per million in the United States
and 14 per million in the UK (Schnakers & Majerus, 2012). The
severity of injury and the cost of hospitalization can cause questions
about end-of-life (Schnakers & Majerus, 2012). High-profile cases,
such as that of Terry Schiavo in 2005, raised awareness that the
consequences of a diagnosis with this population can be far reach-
ing: a diagnosis of “VS” as opposed to MCS in that case resulted in
the decision to withdraw her feeding tube followed by her death
(Schindler v. Schiavo, 2001).
Diagnosis can therefore be seen to influence appropriate care
planning and end-of-life decisions alike. Correct diagnosis is
4 Journal of Music Therapy

important to enable optimal decision-making for disability man-

agement and future care planning, as well as to provide sufficient
family support and to make the best decisions regarding the con-
tinuation of life-preserving interventions (Magee, 2007). A diagno-
sis of “MCS” or “Emerging” requires the person with PDOC to dem-
onstrate that he/she has awareness of his/her environment and is
able to respond purposefully and differently to contrasting stimuli.

Issues in Diagnosis of PDOC

Diagnosis in PDOC remains a complicated matter (Gill-Thwaites,
2006) despite being a primary concern and aim of clinical interven-
tion, given the unacceptably high rates of misdiagnosis, which are
estimated at between 30 and 40% (Hirschberg & Giacino, 2011).
As consciousness cannot be directly observed, the assessment of
populations with PDOC relies on a number of factors, including
rigorous methods of assessment, practitioner experience and skill,
and stability of the patient’s status (Seel et  al., 2010). Behavioral
assessment of the PDOC patient relies on careful observations of
the person’s behaviors at rest (when there is no stimulation), and
comparing these with behaviors that occur during the presentation
of carefully controlled stimuli. As spontaneous and non-purpose-
ful behaviors are typical of the population, repeated measures are
required in order to detect patterns of responsiveness that might
indicate awareness.
Functional MRI scanning may provide greater objectivity in meas-
ures of awareness (Coleman et  al., 2009). However, uncertainty
about the accuracy of diagnoses resulting from such tests due to
technical difficulties confirms behavioral assessments as the pre-
ferred method (Laureys & Schiff, 2012). Thus, much of the research
with PDOC populations focuses on developing standardized meas-
ures for diagnostic assessment, outcome prediction, interdiscipli-
nary treatment planning and projection of future needs, and moni-
toring treatment effectiveness (Giacino, Kalmar, & Whyte, 2004).
Developing reliable and valid behavioral measures for assess-
ment is challenging due to the severity of disability typical of the
population: a combination of motor, sensory, and cognitive impair-
ments can mask residual functioning, thus risking misdiagnosis.
The auditory modality has been the focus of greater empirical
inquiry due to evidence that it is the more sensitive modality for
Vol. 53, No. 1 5

identifying awareness (Gill-Thwaites & Munday, 1999; Owen et al.,

2005), provided by clinical reports of persons with DOC who have
demonstrated cognitive capacity through the auditory modality
even in the absence of movement and language (Giacino et  al.,
2009; Owen et al., 2006). Despite the increased evidence for using
auditory stimuli to assess awareness, the existing standardized
measures for DOC populations fail to address auditory respon-
siveness adequately (Lichtensztejn, Macchi, & Lischinsky, 2014;
Magee, Siegert, Daveson, Lenton-Smith, & Taylor, 2014). The rec-
ommended evidence-based measures reduce auditory stimuli to
verbal material or stimuli that cause a startle response (e.g., blow-
ing a whistle, clapping hands, banging wooden blocks) or involve
calling the patient’s name (Seel et al., 2010).

Music Used for Diagnosis in PDOC

Music is receiving increasing attention as a stimulus for assess-
ment of awareness given its properties of being a non-language-
based stimulus within the auditory modality that has emotional
saliency (Boyle & Greer, 1983; Formisano et al., 2001; Jones et al.,
2000; Lichtensztejn et al., 2014; Magee, 2005, 2007; Magee et al.,
2014; O’Kelly et al., 2013; Okumura et al., 2014; Perrin et al., 2006;
Puggina, da Silva, & Santos, 2011; Verger et al., 2014; Verville et al.,
2012; Wilson, Cranny, & Andrews, 1992). Neurophysiological stud-
ies have revealed that music can promote behavioral responses
indicative of arousal and attention in PDOC patients, and most
importantly, that significant cortical and autonomic nervous system
activity suggest discriminative ability rather than reflex responses
(O’Kelly et al., 2013). Furthermore, music may not just be a useful
tool for evaluation of consciousness, but may stimulate brain activa-
tion that can predict patient improvement (Okumura et al., 2014).
Despite its promise as a diagnostic and prognostic tool, robust
measures and sensitive protocols are needed for using music-
based assessments and interventions in PDOC. Some music thera-
pists working with PDOC populations currently use standardized
neuropsychological measures to assess patient improvements in
response to music therapy (Lichtensztejn et  al., 2014). However,
such measures do not provide adequate protocols for testing
auditory responsiveness and are not sufficiently specific about
responses to auditory stimuli (Lichtensztejn et  al., 2014). This
6 Journal of Music Therapy

misses an opportunity to formulate patient-centered recommenda-

tions for individuals’ rehabilitation and care. Drawing on the evi-
dence underpinning music as a prognostic and diagnostic medium
in PDOC, a measure for assessing awareness in PDOC that involves
the manipulation of music-based stimuli in the auditory modality
was developed to provide a more sensitive assessment of auditory
responsiveness, the Music Therapy Assessment Tool for Awareness
in Disorders of Consciousness (MATADOC).

Addressing Evidence-Based Requirements for Measures for DOC

A number of recommendations for evidence-based behavio-
ral measures utilized with DOC populations have been specified
(Seel et  al., 2010) and were addressed within the current study.
First, we provided specialist training, which included a compre-
hensive manual and clearly defined protocol (Magee, Lenton-
Smith, & Daveson, 2012) to enable consistent administration and
scoring of items, and to provide interpretive guidelines that fit
the Aspen Workgroup consensus-based diagnostic classifications
(Giacino et al., 2002). Second, to address content validity, behav-
iors outlined in the MATADOC are representative of VS, MCS, and
emerging behaviors as outlined in the Aspen Workgroup criteria
(Giacino et al., 2002). Third, we addressed factors for establishing
reliability by ensuring that assessors had sufficient clinical experi-
ence (0.25–19  years, mean  =  3.5  years) with PDOC populations,
and all assessors were trained to a specified level of competency
in using the MATADOC. In addition, when examining test-retest
reliability, change in patient status between scale administrations
was minimized through the study’s design by using video records.
The use of video records allowed us to address the characteris-
tic progression in state of consciousness for PDOC populations.
Fourth, criterion validity was examined through a comparison of
the MATADOC to a standardized reference standard, the SMART
(Gill-Thwaites & Munday, 1999), which has been evaluated as hav-
ing good content validity and may be used to assess DOC with
moderate reservations (Seel et al., 2010). Fifth, construct validity
based on the Rasch model has been examined for the MATADOC
Principal Subscale, which is the only subscale that carries diagnos-
tic utility (Magee et al., 2014). MATADOC subscales two and three
have utility primarily for goal setting and intervention planning.
Vol. 53, No. 1 7

The Rasch analysis indicated that all five items of the Principal
Subscale demonstrated good fit to the Rasch model for the individ-
ual items and the overall model fit. Both principal component and
Rasch analyses of the Principal Subscale, therefore, demonstrate
a robust unidimensional and homogeneous subscale for assessing
awareness in patients with PDOC. The final criterion, that a scale
should predict outcome, remains unknown at the current time.

Purpose of the Current Study

Given a previous study to explore the reliability and dimension-
ality of the MATADOC Principal Subscale (Magee et al., 2014), the
purpose of this study was
i) to examine the reliability (interrater and test-retest) of sub-
scales two and three of the MATADOC, including those items
that previously had not been examined; and
ii) to explore the internal consistency of the two items in the
second subscale that use binary rating for a group of behav-
iors given their clinical rather than diagnostic utility.

Method
Design
We employed a prospective study using repeated measures to
test the reliability of subscales two and three and the internal con-
sistency of the second subscale.

Description and Purpose of the MATADOC

The MATADOC is a 14-item measure that was developed and
refined over a 17-year period for assessment of responsiveness in
adults with PDOC. It is therefore clinically derived with an empha-
sis on its clinical utility within interdisciplinary assessment and
treatment of adults with PDOC following acquired profound brain
injury. Its purpose is to provide a rigorous and detailed assessment
of auditory responsiveness in PDOC patients, given the evidence
that the auditory modality is more sensitive to detecting awareness
in this population (Gill-Thwaites & Munday, 1999; Owen et  al.,
2005). In addition to the auditory modality, it also examines other
behavioral domains that are tested as standard using a protocolized
set of procedures (Magee et  al., 2012). The MATADOC’s overall
8 Journal of Music Therapy

purpose is to demonstrate whether the person with PDOC is aware

by measuring whether he or she responds differently and purpose-
fully to contrasting stimuli and to guide intervention planning
accordingly.
The MATADOC has three subscales that each serve differ-
ent purposes and are described here briefly, followed by a more
detailed description of each subscale (see Table 3). The “Principal
Subscale” tests and reports on responsiveness across several behav-
ioral domains that are included in all DOC measures and con-
tribute to a diagnosis. The second subscale, “Musical Parameter
and Behavioral Response Type,” reports on musical behaviors to
inform music therapy intervention planning. The third subscale,
“Clinical Information to Inform Goal Setting and Clinical Care,”
assesses responses that can be considered indicative of cognitively
mediated responses.

Principal Subscale: Essential Categories.

The Principal Subscale has a total of five items, and assesses
the following five areas: responses to visual stimuli, responses to
auditory stimuli, awareness of musical stimuli, responses to ver-
bal commands, and arousal. Two items within this subscale assess
auditory responsiveness: one rates responses to auditory stimuli
more generally, for example to voices, to single musical sounds,
as well as to more complex musical stimuli. The second auditory
item rates responses to musical stimuli more specifically, such as
songs sung using the patient’s name and familiar music that is
known to be personally meaningful to the patient. The Principal
Subscale is most closely aligned to other validated assessment meas-
ures for PDOC, given that it tests the patient across the auditory,
visual, and communication domains as well as wakefulness (Seel
et al., 2010). These are considered “essential” domains to assess in
DOC measures. The psychometric properties and reliability of this
subscale have been reported previously as having good interrater
reliability (mean = 0.83, sd = 0.11) and good test-retest reliability
(mean = 0.82, sd = 0.05) (Magee et al., 2014). Internal consistency
for the five-item Principal Subscale was rated as good (α = 0.76) for
a four–six-item scale for DOC populations (Seel et al., 2010) with
a strong first principal component. Rasch analysis confirmed these
impressions of a reliable, unidimensional, and homogeneous scale.
Vol. 53, No. 1 9

The MATADOC Principal Subscale has been compared with

an external reference standard to test its concurrent validity. The
Sensory Modality Assessment and Rehabilitation Technique, or
SMART (Gill-Thwaites, 1997), is an established and standardized
global measure for PDOC that assesses responsiveness across sen-
sory modalities (Gill-Thwaites & Munday, 1999). Excellent agree-
ment (100%) was observed between MATADOC and SMART out-
comes for diagnosis of awareness states, with Cohen’s κ = 1 providing
strong evidence of validity and Fisher’s exact test demonstrating a
strong association (p < 0.001) (Magee et al., 2014). However, con-
trasting sensitivities between the measures have been highlighted,
with MATADOC having a higher sensitivity within the auditory and
visual domains relative to SMART, and SMART having a higher sen-
sitivity in the motor domain (O’Kelly & Magee, 2013). These find-
ings indicated that both measures have contrasting sensitivities,
providing data predictive of awareness from mutually exclusive
behavioral domains: SMART and MATADOC may elicit different
levels of response in comparable domains. The musical response
items in MATADOC provide distinctive behavioral data that may
contribute to prediction of awareness. These findings highlight
the importance of concurrent use of varied measures to provide
a comprehensive multisensory assessment of PDOC patients, sup-
porting the recommendations of Giacino et al. (2002). Although
the MATADOC has demonstrated excellent concurrent validity
with the SMART, its predictive validity remains untested.

Second Subscale: Musical Parameter and Behavioral

Response Type.
The second subscale has a total of two items, and assesses two
areas (Behavioral Response to Music; Musical Response) that
inform intervention planning. Item 6: Behavioral Response to
Music rates behaviors contingent to musical stimuli across six sub-
items, such as wakefulness, physical movement, facial gesture, res-
piration rate and eye direction, and vocalization. Item 7: Musical
Response has seven sub-items that rate patient responsiveness to
specific musical parameters (pulse/rhythm, melody/pitch, tim-
bre, dynamics/intensity, form, tempo, and musical mood). The
purpose of this subscale is to provide detailed information about
responsiveness to specific components of the auditory environment
10 Journal of Music Therapy

and any behaviors that are contingent to stimuli. For example,

responses to manipulation of volume can provide important infor-
mation for managing the patient’s auditory environment; contin-
gent responses to manipulation of timbre (e.g., change of instru-
ment) or musical form (e.g., pauses in the music) can provide
evidence of purposeful response that can contribute to a diagnosis
of awareness. Detailed information of this type helps both tailor
individualized music therapy intervention optimally suited to an
individual patient’s responsiveness as well as provide information
to assist carers in structuring the patient’s auditory environment.
This is important for people who are fully dependent on others for
all aspects of care.

Third Subscale: Clinical Information to Inform

Goal Setting and Clinical Care.
The third subscale has a total of seven items, and assesses motor,
communication, and emotional behaviors. This subscale serves to
inform goal setting and clinical care. Two items examine communi-
cative behaviors drawing on musical behaviors: nonverbal commu-
nication and vocalization. Thus, six items of the entire MATADOC
measure focus on auditory responsiveness. The seven items of
the third subscale extend beyond the essential behaviors that are
assessed in PDOC (e.g., visual responsiveness, wakefulness) and
that are covered in the Principal Subscale. All of the items in the
third subscale rate behaviors that are associated with cognitively
mediated behaviors, for example Item 10: Choice-Making or Item
13: Intentional Behaviour. Thus, the items are well suited for pro-
cedures that develop specific patient strengths as part of interdisci-
plinary goal-oriented rehabilitation.

The MATADOC Protocol

The MATADOC protocol uses a wide range of musical stimuli,
including isolated single auditory stimuli (i.e., a single pitch on
a single instrumental timbre), complex musical sounds (e.g., the
strum of a guitar involving multiple pitches on a single instrumen-
tal timbre; singing a familiar song with keyboard accompaniment
enabling the manipulation of instrumental timbre, pitch, har-
mony, and volume), and musical activities to measure a number of
functional behaviors across the motor, communication, visual, and
Vol. 53, No. 1 11

auditory domains (e.g., goal-directed movement to make a sound

on an instrument; use of vocalization within a song) (Magee et al.,
2012).
At its simplest, the MATADOC protocol involves a minimum of
five procedures that aim to elicit patient responses. Depending on
responsiveness, the five procedures can be extended to assess and
develop, where possible, more complex behaviors such as pragmatic
and functional communication, motor responses, or cognitively
mediated functions such as choice-making and attention to task.
All procedures use music or music-related stimuli. The music used
is predominantly presented live, although recorded music can be
used within one procedure for genres that are known to be person-
ally meaningful to the patient but are difficult to recreate convinc-
ingly by an assessor (e.g., opera, hip hop, world). The music used
varies between novel unfamiliar music and salient music known to
have personal meaning to the patient. Personally meaningful music
is determined through a prior interview with family. The assessment
protocol is followed and administered over the course of four indi-
vidual assessment sessions, lasting between 15 and 30 minutes each
depending on patient tolerance, within a ten-day period. A  short
behavioral observation period before and after each assessment ses-
sion allows for data collection of “at-rest” behaviors.

Scoring the MATADOC

The MATADOC provides three scores that contribute to a diag-
nosis (see Table 3). After rating the observed behaviors across 14
items, the assessor converts the ratings into a numerical score for
the Principal and second subscales. The third subscale converts rat-
ings of each of its items to an associated behavioral category of “VS,
MCS, or Emerging.”.
The score produced by the Principal Subscale has diagnostic
utility (Magee et al., 2014). The score provided by subscale two pri-
marily has clinical utility. It uses binary rating (yes/no) to identify
which musical components elicit observed behavioral responses,
thereby assisting the music therapist with planning intervention.
The binary scoring used in subscale two differentiates it from the
other subscales. The scored outcomes in subscale three (VS, MCS,
or Emerging for each item) support the diagnostic score provided
by the Principal Subscale.
12 Journal of Music Therapy

Participants
A convenience sample was recruited from a specialist unit
for adults with PDOC who were admitted to the facility for the
purposes of gaining an accurate diagnosis through interdisci-
plinary assessment. Participants were required to have no con-
firmed diagnosis of VS, MCS, or Emerging at the time of recruit-
ment. Participants were required to be medically stable and be
aged between 16 and 70  years of age. Participants with known
pre-morbid hearing impairments, a previous diagnosis of musi-
cogenic epilepsy, or who emerged from DOC during recruitment
or assessment were excluded. Mental capacity assessments were
completed before recruitment, and standard procedures regard-
ing the recruitment of patients lacking capacity were followed.
Ethics was gained from the UK National Research Ethics Service
(05/Q0406/185). All participants received standard interdiscipli-
nary rehabilitation assessment and intervention during the study.

Procedures for Data Collection

Data were collected for each participant during four assessment
sessions within a 10-day period. The MATADOC protocol that has
previously been described (Magee et  al., 2012) was followed in all
assessment sessions. Two independent assessors were involved in
every assessment session (Assessors A  and B). Assessor A  delivered
the protocol and rated behaviors, and Assessor B observed and rated
behaviors, thus the same live assessment session was rated simultane-
ously by two independent assessors. A paper form of the MATADOC
assessment was completed by each assessor at the end of each assess-
ment session. The four live assessment sessions were video recorded,
and the video records were rated again by the same two assessors using
paper forms of the MATADOC at a later date for test-retest analysis.
Three-minute observational periods were conducted at rest
immediately before and after each MATADOC session in line
with best evidence-based practice recommendations for DOC
(Seel et al., 2010). Behavioral data collected during these observa-
tions did not contribute to scoring. Behavioral observations with
no stimulation are essential to determine whether the behaviors
observed during musical stimuli are purposeful, as the purpose
of DOC measures is to determine whether the person with PDOC
responds differently and purposefully to contrasting stimuli.
Vol. 53, No. 1 13

Procedures to Minimize Bias
All the assessors in this study were credentialed music therapists
who were trained in using the MATADOC. Seven assessors in total
were involved in data collection, whose experience with PDOC
populations ranged from 0.25 to 19 years (mean = 3.5 years). The
two assessors involved with each individual participant remained
consistent across that individual’s entire MATADOC assessment
(four live clinical contacts and four video ratings). Assessors were
blinded to each other’s ratings for interrater analysis. An adequate
period of time (between four and 24 weeks) elapsed between the
live session and the video observation to minimize the assessor’s
memory of the previous session. Furthermore, viewing order of
the clinical contacts was randomized, providing further interfer-
ence with any memory of the live MATADOC ratings. All data were
anonymized and entered into an EXCEL spreadsheet by a data
manager who was independent of this project. Later, data were
imported into SPSS for statistical analysis.

Analysis
Interrater and test-retest reliability were assessed using the
intraclass  correlation coefficient with random effects, commonly
referred to as ICC(2), to account for there being multiple raters
(n = 7). Interrater reliability analysis used only live assessment ses-
sions but included all repeated contacts (n = 4 each), which were
assumed to be independent. This resulted in a sample size of 168
(21 x 4 x 2) observations for interrater analysis. A similar type of
assumption was made for test-retest. Test-retest reliability analysis
compared ratings of live and video observations of the same con-
tact undertaken by the same assessor (168 ratings: 21 x 4 x 2). The
intra-patient variation across multiple contacts in a PDOC popula-
tion is sufficient to support the use of this technique (see Figure 1).
The two items contained within the second subscale (MATADOC
items 6 and 7) were examined independently for internal consist-
ency. This was necessary, given that these items were structured and
scored differently to the items in the Principal and third subscales
using binary rating. This subscale warranted special examination,
given its primary utility to inform music-specific intervention plan-
ning and the arbitrary nature of its numerical scoring: “no” was
scored 1, and “yes” was scored 2.  Internal consistency of scores
14 Journal of Music Therapy

Figure 1.
Matrix of live and video ratings to achieve inter-rater and test-retest ratings.

derived from multiple responses was assessed for items 6 and 7

using Cronbach’s alpha, using live observations only (excluding
video observations) and assuming the correlations of items do not
change over repeated assessments of the same patients. All analyses
were conducted using R version 3.1.2 (R Core Team, 2014) and the
“psych” package (Revelle, 2014).
Criteria for assessing measures for rehabilitation vary signifi-
cantly in the literature. Although recommendations propose a rat-
ing scheme for interrater and test-retest reliability in DOC scales
where coefficients of < 0.70 are unacceptable (Seel et  al., 2010),
others have stressed that measures for disability should balance tra-
ditional psychometric and clinimetric considerations of such meas-
ures: criteria have been proposed where coefficients of > 0.40 to
< 0.75 are rated as adequately covered but with weakness in some
domains or lacking in a noncentral domain (Andresen, 2000).
Given the primary purposes of the two subscales under exami-
nation being for clinical rather than diagnostic utility, we have
adopted the recommendations laid forth by Andresen (2000).

Results
Sample Characteristics
Over a 36-month period, twenty-one participants (11 male;
10 female) were recruited with a mean age of 40.3  years
(range = 19–67; SD = 15.65). All participants had profound brain
Vol. 53, No. 1 15

damage and were completely dependent for all activities of daily

living, with profound communication impairments and unknown
sensory responsiveness. Etiology of brain injury covered trauma
(n  =  9), hypoxic/ischaemic (n  =  7), haemorrhagic (n  =  3), and
viral infection (n = 2). The average time since onset was 7.9 months
(see Table 1 for total characteristics). MATADOC outcomes for the
sample ranged across VS (n = 13) and MCS (n = 8). There were no
cases who were diagnosed as Emergent. These diagnoses were in
100% agreement with diagnoses provided by an external reference
standard (see Magee et al., 2014).

Reliability
We report in detail here on subscales two and three (items 6–14)
of the MATADOC, as IRR and TRT for the Principal Subscale
have already been reported (Magee et al., 2014). Calculated intra-
class  correlation coefficients (ICCs) for items from subscales two
and three of the MATADOC produced mixed results. ICCs for
Table 1
Patient Cohort Recruited to the Study

Patient number Gender Age Etiology of brain damage Time since onset (months)

1 M 21 Traumatic 7
2 M 25 Traumatic 8
3 M 60 Hypoxic/ischaemic 5
4 M 23 Traumatic 10
5 M 19 Hypoxic/ischaemic 9
6 F 42 Hypoxic/ischaemic 8
7 F 19 Traumatic 13
8 M 48 Haemorrhagic 5
9 F 27 Viral 15
10 F 37 Traumatic 6
11 F 35 Hypoxic/ischaemic 6
12 F 47 Haemorrhagic 8
13 F 59 Traumatic 5
14 M 36 Hypoxic/ischaemic 6
15 M 65 Traumatic 16
16 F 58 Viral 10
17 M 44 Traumatic 7
18 M 45 Hypoxic/ischaemic 5
19 M 23 Traumatic 6
20 F 67 Hypoxic/ischaemic 5
21 F 46 Haemorrhagic 5
16 Journal of Music Therapy

IRR ranged from 0.32 to 0.74 (mean = 0.48) for seven of the nine
items, with a mean ICC of 0.41 for the two subscales overall. Item
7: Musical Response and Item 10: Choice-Making had much lower
ICCs, 0.14 and 0.17, respectively. Using the criteria proposed by
Andresen (2000), Items 8–9 and 11–12 had adequate IRR but
Items 7, 10, and 14 had poor IRR, with items 6 and 13 also falling
just inside the threshold for poor IRR (0.39).
Test-retest reliability for most of the items fell within the accept-
able range (range  =  0.55–0.69, mean  =  0.61) with exception of
Item 7: Musical Response (0.26) (see Table 2). The results indicate
acceptable TRR for Items 6 and 8–14. However, Item 7: Musical
Response has poor TRR.

Internal Consistency
Items 6 and 7 were examined independently for internal con-
sistency, given that they each have nonhierarchical sub-items
with binary rating. They were examined as separate rather than
combined items, as they do not examine the same construct.
Cronbach’s alpha for Item 6: Behavioural Response (with six sub-
items) was 0.51; for Item 7: Music Response (seven sub-item), it
was 0.28.These results indicate poor internal consistency for Items
6 and 7 (See Table 2).

Discussion
This study aimed to examine the reliability of the two subscales of
the MATADOC, “Subscale Two: Musical Parameter and Behavioural
Response Type” and “Subscale Three: Clinical Information to
Inform Goal Setting and Clinical Care.” These subscales were
examined separately, as they were believed to hold greater clini-
cal utility rather than psychometric strength. This contrasts with
the MATADOC Principal Subscale, which holds diagnostic utility
and so was tested for its psychometric properties separately. The
three subscales have combined strength as a comprehensive meas-
ure that can rate responsiveness to music-based auditory stimuli
in PDOC populations. Thus, this study aimed to provide a clearer
picture of the MATADOC’s overall psychometrics.
Testing for IRR and TRT for subscales two and three of the
MATADOC entire measure revealed mixed results. Four of the items
had adequate reliability, with a further two reaching just below the
 Vol. 53, No. 1

Table 2
Descriptive Statistics (N = number of pairs of observations, M = mean score, SD = standard deviation), Internal Consistency (Cronbach’s α) of Derived
Scores, and Interrater and Test-Retest Reliability (estimated using intraclass correlation coefficients, ICC) for MATADOC Subscales 2 and 3

Internal consistency Interrater reliability Test-retest reliability

Variable N M SD α 95% CI ICC 95% CI ICC 95% CI

Score Q6, items summed 82 1.40 1.33 0.51 (0.37, 0.65) 0.39 (0.17, 0.54) 0.62 (0.52, 0.70)
Score Q7, items summed 84 0.36 0.67 0.28 (0.10, 0.46) 0.14 (–0.08, 0.34) 0.26 (0.11, 0.39)
Vocalization 84 –0.01 0.83 0.74 (0.63, 0.82) 0.60 (0.49, 0.69)
Nonverbal communication 84 0.37 0.75 0.63 (0.48, 0.74) 0.63 (0.53, 0.72)
Choice-making 84 0.08 0.32 0.17 (–0.04, 0.37) 0.58 (0.47, 0.67)
Motor skills 84 0.58 0.86 0.47 (0.25, 0.59) 0.61 (0.50, 0.69)
Attention to task 84 0.45 0.57 0.43 (0.21, 0.57) 0.61 (0.50, 0.69)
Intentional behaviour 84 0.52 0.83 0.39 (0.18, 0.55) 0.69 (0.60, 0.76)
Emotional response 84 0.57 1.02 0.32 (0.12, 0.50) 0.55 (0.43, 0.64)
17
Table 3
18

MATADOC Subscales, Items and Their Purpose, and Score Derivations

Subscale title/Items Purpose Method of score derivation

Principal Subscale (Essential categories)

1: Responses to visual Rates responses to visual stimuli presented without an Ratings are converted to a score of 0–2,
stimuli accompanying auditory component. where 0 is indicative of a “VS” level
2: Responses to auditory Rates responses to auditory stimuli presented without visual stimuli response; 1 indicative of an “MCS”
stimuli for assessing all evidence of residual hearing including startle level response; and, 2 indicative of an
responses. “Emerging” level response.
3: Awareness of musical Rates behavioral and musical responses that provide evidence of The summed total of the five items’ scores
stimuli the patient’s awareness of the music (live or recorded). are linked to a diagnostic category:
4: Response to verbal Rates responses to one-step verbal commands. a) 0–3 = VS
commands b) 4–7 = MCS
5: Arousal Rates arousal across the session demonstrate by eyes open and c) ≥ 8 = Emerging
being alert.

Second subscale: Musical parameter and behavioraul response type

6: Behavioural response Rates all observable cognitively mediated behaviors at the initial Each sub-item has a binary rating: “No”
to music presentation of musical stimulation or throughout other score = 1; “Yes” score = 2. Total score
periods within the assessment session. Records differential range for subscale: 13–26.
responses, i.e., behaviors which occur contingent to musical Sum of scores tracks developments or
stimuli in 6 sub-items, each of which is rated independently. losses in overall responsiveness.
7: Musical response Rates responses to musical parameters in 7 sub-items, each of Holds no diagnostic value but assists
which is rated independently. Useful for collecting information with intervention planning for use of
to inform subsequent music therapy interventions specifically musical stimuli.
tailored to meet the patient’s responses.
Journal of Music Therapy
Table 3 
Continued
Subscale title/Items Purpose Method of score derivation
Vol. 53, No. 1

Third subscale: Clinical information to inform goal setting and clinical care
8: Vocalization Documents the patient’s vocal activity (including reflexive Item ratings are converted to scores that
behaviors and cognitively mediated responses). are categorized as Absent/
9: Nonverbal Rates observed responses pertaining to nonverbal social VS, Developing/MCS, or Strength/
communication communication responses. Emerging responses.
10: Choice-making Rates the ability to discriminate between a forced choice of two Holds no diagnostic value but
alternatives presented using the communication strategy used a) supports the diagnosis formed from
by the treatment team. the Principal Subscale;
11: Motor skills Rates observed physical responses to musical objects, sounds, and b) assists with uni/interdisciplinary goal
music presented. setting, and
12: Attention to task Rates volitional behaviors denoting attention (focused or c) assists with monitoring patient
sustained) during procedures involving musical stimuli, be they responsiveness against the goals that
auditory (e.g., musical sounds) or combined auditory/ have been set.
visual.
13: Intentional behaviour Rates observed goal-oriented behaviors.
14: Emotional response Rates emotionally expressive behaviors that are contingent
to musical stimuli including facial gestures, tears, and
vocalizations.
19
20 Journal of Music Therapy

threshold of “adequate” for IRR. TRT reliability was stronger over-

all for nearly all items. Item 10: Choice-Making had poor IRR, and
Item 7: Musical Response had poor IRR and TRT. Furthermore,
Item 7 and Item 6: Behavioural Response have poor internal con-
sistency. This contrasts with the MATADOC Principal Subscale that
as a five-item scale has good IRR and TRT and good internal con-
sistency (Magee et al., 2014).
Given the poor results for “Item 7: Musical Response,” one option
is to remove this item from the MATADOC. However, as this is the
primary item for rating music-specific parameters that have elic-
ited responsiveness in this complex population, its removal would
be shortsighted. Therefore, examining underlying reasons for its
poor IRR/TRT warrants further explanation. Several factors that
may have contributed to poor IRR/TRT for Item 7 include limited
sensitivity to change due to the scoring system, assessor experience,
and MATADOC training.
First, the sub-items are added into an arbitrary score that has no
real value: the “score” offers a numerical figure based on the clini-
cian’s simple behavioral ratings of “present” or “absent” behaviors.
These ratings can be elaborated in a useful manner to guide deci-
sions in tailoring the intervention for the next assessment session.
So, it is stressed that the scores derived from Item 7 do not contrib-
ute to diagnosis, but remain relevant for planning intervention.
The purpose of Item 7: Musical Response is to guide clinical
decision-making for tailoring intervention to an individual’s spe-
cific responses to musical stimuli. Breaking down auditory respon-
siveness in this way is important to enhance the sensitivity of the
MATADOC assessment; not only does it assist the music therapist
to tailor intervention, but it also guides recommendations for
managing the auditory environment (e.g., in ward environments)
and can contribute to global communication strategies. These
contributions are important for the heavily dependent patient
with profound communication impairment who is reliant on oth-
ers to manage her/his auditory environment (e.g., whether or
not to play music, television, etc., or have a period of no stimula-
tion or silence). However, reducing musical experiences to musi-
cal parameters (e.g. “tempo”; “timbre”) for rating in this way may
be an underlying problem for the psychometric properties of this
item. That is, the variance between the sub-items may just be too
small. Thus, Item 7 is weak psychometrically for reasons of scoring
Vol. 53, No. 1 21

but holds clinical utility, as it focuses the therapist’s attention to

the musical parameters relating to auditory responsiveness in the
PDOC patient.
Second, it could be argued that this item may reveal variance
in the assessors’ skills. Rating responses on this item requires the
therapist/assessor to make decisions as to whether the observed
behaviors are related specifically to the music being played in
the session by the therapist/assessor. For example, the therapist
assesses whether motor movements observed are at a similar tempo
to music being played or reflect the pulse of the music; or whether
client-initiated vocal sounds are at the pitch of the music being
played. Despite training to a recognized level of competency, the
skills required to assess patient responses in this complex popula-
tion are highly variable between assessors. Within the music ther-
apy profession, there is considerable variation as to music-centered
practices, influenced by factors such as the model in which the
therapist trained, the clinical population worked with, and the care
setting worked in. Similarly, there was little consistency within the
research team in this study that comprised seven therapists from six
different training backgrounds. In reality, some therapists will make
more use of Item 7 than others, and use it differently from others,
for example the degree to which a therapist makes links between
(musical) cause and (behavioral outcome) effect. However, remov-
ing the item altogether prevents a music therapist from making a
finely tuned assessment of the patient’s auditory responsiveness to
musical parameters. In essence, it is Item 7: Musical Response that
makes the MATADOC a music-based assessment.
Third, poor IRR and TRT suggest that the training for therapists
to deliver and rate this item is currently inadequate and needs fur-
ther attention. Feedback from follow-up MATADOC training ses-
sions has revealed that trainees are not consistent in their under-
standing of how to deliver protocol procedures relating to this
item, nor of how to assess responses. This is now being addressed
in MATADOC training, using illustrative online videos and greater
explanation of how to assess and rate Item 7 and its relevance for
patient care. Thus, despite the item’s poor psychometric proper-
ties, it has clinimetric value.
Finally, this study involved a relatively small sample of patients
who do not demonstrate much change over time. Consequently,
this sample may not have provided enough variance to properly
22 Journal of Music Therapy

evaluate the psychometric quality of this item. A  sample that

included patients with a greater variance of consciousness (i.e.,
including people diagnosed as “Emerging”) may show Item 7 in a
better light or at least enable improved evaluation.
The internal consistency of both Items 6 and 7 suggests that the
within-item constructs are poorly related. Given the range of con-
structs included in each item, this seems accurate. This is not to
suggest that the items are not useful or unimportant, but simply
that no importance should be given to the arbitrary single total
scores they produce, as this could be analogous to adding apples
and oranges and pears.
Similarly, Item 10: Choice-Making was revealed to be a weaker
item in terms of interrater reliability. Item 10 rates abilities that
concern developing communication skills and using these to com-
municate preference (e.g., for an instrument; for a song; for a
favorite music performer). Thus, it is clear that this item is appro-
priate only for MCS or Emergent patients. The variability we found
can be explained by the difficulty in using a single systematic pro-
cedure to assess for this behavioral domain. “Choice-Making” in
this patient group relies on the patient’s developing communica-
tion skills. Typically with this patient group, a variety of behaviors
emerge to indicate “communication” in the period of the patient’s
recovery when a communication strategy is developing. The com-
plexity of patient response is highly variable (e.g., eye gaze versus
gesture for “yes”) and often inconsistent. In addition, variability in
assessors’ ability and confidence to assess this skill may have con-
tributed to this item being weaker in terms of IRR. As with Item 7,
this has now been addressed in training and is given greater expli-
cation. In reality, this is an advanced clinical practice skill for pro-
fessionals working in PDOC and typically relies on learning from
interdisciplinary colleagues with specialist skills.

Conclusion
Overall, the results need to be considered in light of this measure
being a work in progress. The entire MATADOC offers clinicians a
useful measure for documenting behaviors in a consistent manner.
It demonstrates sensitivity through its capacity to rate behaviors in
small enough increments to reflect change, a necessary function
for evaluating responsiveness in minimally responsive populations.
Vol. 53, No. 1 23

It provides a tool for assessment and evaluation of intervention,

and thus has potential for demonstrating patient change over time.
It has clinical utility due to its ease of use once trained, its adapt-
ability for a range of music therapy practices, and its ability to pro-
vide information that is essential to diagnosis of the PDOC patient.
In a previous study, we established that the Preliminary Subscale
has diagnostic utility, including good IRR, TRT, and internal con-
sistency (Magee et al., 2014). In this study, we established that sub-
scales two and three have IRR and TRT to a degree that suggests the
MATADOC’s utility for guiding clinical intervention. To improve
MATADOC implementation, training now gives greater priority
to three areas: 1)  presenting procedures for musical responsive-
ness assessment, 2) how to assess behaviors that can be deemed as
musical responsiveness, and 3) training assessors in advanced skills
to assess the patient’s communication strategies. Future research
should include a larger and more varied sample that includes VS,
MCS, and Emerging patients to provide improved examination of
the second subscale.

Funding
Funding for this project was made possible by the Neuro-Disability
Research Trust, Royal Hospital for Neuro-Disability, London, and
the Society for Education, Music, and Psychology Research, UK.

Acknowledgements
The authors would like to acknowledge the following at the
Royal Hospital for Neuro-Disability, London: the Departments of
Music Therapy, Speech and Language Therapy, and Occupational
Therapy, which contributed to the data collection; the Research
Department; Eirini Alexiou, Senior Music Therapist, who made a
significant contribution to data collection; and Maura Quinn, who
provided data management.

References
Andresen, E. M. (2000). Criteria for assessing the tools of disability outcomes
research. Archives of Physical Medicine and Rehabilitation, 81, S15–S20.
Boyle, M. E., & Greer, R. D. (1983). Operant procedures and the comatose patient.
Journal of Applied Behavior Analysis, 16(1), 3–12.
24 Journal of Music Therapy

Bruno, M., Vanhaudenhuyse, A., Thibaut, A., Moonen, G., & Laureys, S. (2011).
From unresponsive wakefulness to minimally conscious PLUS and func-
tional locked-in syndromes: Recent advances in our understanding of disor-
ders of consciousness. Journal of Neurology, 258(7), 1373–1384. doi:10.1007/
s00415-011-6114-x
Coleman, M. R., Davis, M. H., Rodd, J. M., Robson, T., Ali, A., Owen, A. M., &
Pickard, J. D. (2009). Towards the routine use of brain imaging to aid the clini-
cal diagnosis of disorders of consciousness. Brain: A Journal of Neurology, 132,
2541–2552. doi:10.1093/brain/awp183
Formisano, R., Vinicola, V., Penta, F., Matteis, M., Brunelli, S., & Weckel, J. W. (2001).
Active music therapy in the rehabilitation of severe brain injured patients dur-
ing coma recovery. Annali Dell’Istituto Superiore Di Sanità, 37(4), 627–630.
Giacino, J. T., Ashwal, S., Childs, N., Cranford, R., Jennett, B., Katz, D. I., … Zasler,
N. D. (2002). The minimally conscious state: Definition and diagnostic criteria.
Neurology, 58(3), 349–353.
Giacino, J. T., Kalmar, K., & Whyte, J. (2004). The JFK coma recovery scale-revised:
Measurement characteristics and diagnostic utility. Archives of Physical Medicine
and Rehabilitation, 85(12), 2020–2029.
Giacino, J. T., Schnakers, C., Rodriguez-Moreno, D., Kalmar, K., Schiff, N., & Hirsch,
J. (2009). Behavioral assessment in patients with disorders of consciousness:
Gold standard or fool’s gold? Progress in Brain Research, 117, 33–47.
Gill-Thwaites, H. (1997). The sensory modality assessment rehabilitation technique—
A tool for assessment and treatment of patients with severe brain injury in a
vegetative state. Brain Injury, 11(10), 723–734. doi:10.1080/026990597123098
Gill-Thwaites, H. (2006). Lotteries, loopholes, and luck: Misdiagnosis in the vegeta-
tive state patient. Brain Injury, 20(13–14), 1321–1328.
Gill-Thwaites, H., & Munday, R. (1999). The sensory modality assessment and reha-
bilitation technique (SMART): A  comprehensive integrated assessment and
treatment protocol for the vegetative state and minimally responsive patient.
Neuropsychological Rehabilitation, 9(3–4), 305–320. doi:10.1080/096020199389392
Hirschberg, R., & Giacino, J. T. (2011). The vegetative and minimally conscious
states: Diagnosis, prognosis, and treatment. Neurologic Clinics, 29(4), 773–786.
doi:10.1016/j.ncl.2011.07.009
Jennett, B., & Plum, F. (1972). Persistent vegetative state after brain damage: A syn-
drome in search of a name. Lancet, 1(7753), 734–737.
Jones, S. J., Pato, M. V., Sprague, L., Stokes, M., Munday, R., & Haque, N. (2000).
Auditory evoked potentials to spectro-temporal modulation of complex
tones in normal subjects and patients with severe brain injury. Brain, 123(5),
1007–1016.
Laureys, S., Celesia, G. G., Cohadon, F., Lavrijsen, J., León-Carrión, J., Sannita,
W. G., … the European Task Force on Disorders of Consciousness. (2010).
Unresponsive wakefulness syndrome: A  new name for the vegetative state or
apallic syndrome. BMC Medicine, 8, 68–71. doi:10.1186/1741-7015-8-68
Laureys, S., & Schiff, N. D. (2012). Coma and consciousness: Paradigms (re)
framed by neuroimaging. Neuroimage, 61(2), 478–491. doi:10.1016/j.
neuroimage.2011.12.041
Lichtensztejn, M., Macchi, P., & Lischinsky, A. (2014). Music therapy and disorders
of consciousness: Providing clinical data for differential diagnosis between
Vol. 53, No. 1 25

vegetative state and minimally conscious state from music-centered music

therapy and neuroscience perspectives. Music Therapy Perspectives, 32(1), 47–55.
doi:10.1093/mtp/miu001
Magee, W. L. (2005). Music therapy with patients in low-awareness states: Approaches
to assessment and treatment in multidisciplinary care. Neuropsychological
Rehabilitation, 15(3–4), 522–536. doi:10.1080/09602010443000461
Magee, W. L. (2007). Music as a diagnostic tool in low-awareness states: Considering
limbic responses. Brain Injury, 21(6), 593–599.
Magee, W. L., Lenton-Smith, G., & Daveson, B. A. (2012). Music Therapy Assessment
for Awareness in Disorders of Consciousness (MATADOC): Assessment Manual and
Instructions for Use. London: Royal Hospital for Neuro-Disability
Magee, W. L., Siegert, R. J., Daveson, B. A., Lenton-Smith, G., & Taylor, S. M. (2014).
Music therapy assessment tool for awareness in disorders of consciousness
(MATADOC): Standardisation of the principal subscale to assess awareness in
patients with disorders of consciousness. Neuropsychological Rehabilitation, 24(1),
101–124. doi:10.1080/09602011.2013.844174
O’Kelly, J., James, L., Palaniappan, R., Taborin, J., Fachner, J., & Magee, W. L.
(2013). Neurophysiological and behavioral responses to music therapy in vegetative
and minimally conscious states. Switzerland: Frontiers Research Foundation.
doi:10.3389/fnhum.2013.00884
O’Kelly, J., & Magee, W. L. (2013). The complementary role of music therapy in the
detection of awareness in disorders of consciousness: An audit of concurrent
SMART and MATADOC assessments. Neuropsychological Rehabilitation, (23)2,
287–298.
Okumura, Y., Asano, Y., Takenaka, S., Fukuyama, S., Yonezawa, S., Kasuya, Y., &
Shinoda, J. (2014). Brain activation by music in patients in a vegetative or
minimally conscious state following diffuse brain injury. Brain Injury, 28(7),
944–950. doi:10.3109/02699052.2014.888477
Owen, A. M., Coleman, M. R., Boly, M., Davis, M. H., Laureys, S., & Pickard, J. D.
(2006). Detecting awareness in the vegetative state. Science, 313(5792), 1402.
Owen, A. M., Coleman, M. R., Menon, D. K., Johnsrude, I. S., Rodd, J. M., Davis,
M. H., … Pickard, J. D. (2005). Residual auditory function in persistent veg-
etative state: A combined PET and fMRI study. Neuropsychological Rehabilitation,
15(3–4), 290–306. doi:10.1080/09602010443000579
Perrin, F., Schnakers, C., Schabus, M., Degueldre, C., Goldman, S., Bredart, S.,
… Laureys, S. (2006). Brain response to one’s own name in vegetative state,
minimally conscious state, and locked-in syndrome. Archives of Neurology, 63,
562–569.
Plum, F., and Posner, J. B. (1983). The diagnosis of stupor and coma (3rd ed.).
Philadelphia: FA Davis.
Puggina, A. C. G., da Silva, M. J. P., & Santos, J. L. F. (2011). Use of music and voice
stimulus on patients with disorders of consciousness. Journal of Neuroscience
Nursing, 43(1), E8–E16.••
R Core Team. (2014). R: A  language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. Retrieved from http://
www.R-project.org/.
Revelle, W. (2014). psych: Procedures for Personality and Psychological Research.
Northwestern University, Evanston, IL.
26 Journal of Music Therapy

Royal College of Physicians. (2013). Prolonged disorders of consciousness. National clini-

cal guidelines. London.
Schindler v. Schiavo, 9–10 Florida Sixth Judicial Circuit. (2001).
Schnakers, C., & Majerus, S. (2012). Behavioral assessment and diagnosis of disor-
ders of consciousness. In C. Schnakers, & S. Laureys (Eds.), Coma and disorders
of consciousness (pp. 1–10). New York: Springer Science + Business Media.
Seel, R. T., Sherer, M., Whyte, J., Katz, D. I., Giacino, J. T., Rosenbaum, A. M., …
Zasler, N. (2010). Assessment scales for disorders of consciousness: Evidence-
based recommendations for clinical practice and research. Archives of Physical
Medicine and Rehabilitation, 91(12), 1795–1813. doi:10.1016/j.apmr.2010.07.218
Verger, J., Ruiz, S., Tillmann, B., Ben Romdhane, M., De Quelen, M., Castro, M., …
Perrin, F. (2014). Beneficial effect of preferred music on cognitive functions
in minimally conscious state patients. Revue Neurologique, 170(11), 693–699.
doi:10.1016/j.neurol.2014.06.005
Verville, V. C., Sela, L., Plotkin, A., Chatelle, C., Sobel, N., & Laureys, S. (2012).
Detecting signs of consciousness in severely brain injured patients with vol-
untary control of sniffing: A  cohort study. Accepted Abstracts from the
International Brain Injury Association’s Ninth World Congress on Brain Injury.
Brain Injury, 26 (4–5), 721.
Wilson, S. L., Cranny, S. M., & Andrews, K. (1992). The efficacy of music for stimu-
lation in prolonged coma: Four single-case experiments. Clinical Rehabilitation,
6, 181–187.
Copyright of Journal of Music Therapy is the property of Oxford University Press / USA and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.