Review

Music-based interventions in neurological rehabilitation

Aleksi J Sihvonen, Teppo Särkämö, Vera Leo, Mari Tervaniemi, Eckart Altenmüller, Seppo Soinila

Lancet Neurol 2017; 16: 648–60 During the past ten years, an increasing number of controlled studies have assessed the potential rehabilitative effects
Published Online of music-based interventions, such as music listening, singing, or playing an instrument, in several neurological
June 26, 2017 diseases. Although the number of studies and extent of available evidence is greatest in stroke and dementia, there is
http://dx.doi.org/10.1016/
also evidence for the effects of music-based interventions on supporting cognition, motor function, or emotional
S1474-4422(17)30168-0
wellbeing in people with Parkinson’s disease, epilepsy, or multiple sclerosis. Music-based interventions can affect
Faculty of Medicine
(A J Sihvonen MD) and divergent functions such as motor performance, speech, or cognition in these patient groups. However, the
Department of Neurology psychological effects and neurobiological mechanisms underlying the effects of music interventions are likely to
(S Soinila MD), University of share common neural systems for reward, arousal, affect regulation, learning, and activity-driven plasticity. Although
Turku, Turku, Finland;
further controlled studies are needed to establish the efficacy of music in neurological recovery, music-based
Cognitive Brain Research Unit,
Department of Psychology and interventions are emerging as promising rehabilitation strategies.
Logopedics, Faculty of
Medicine, University of Introduction controlled trials investigating the effects of music-based
Helsinki, Finland (A J Sihvonen,
The world’s population is ageing rapidly and the number interventions in the rehabilitation of patients with
T Särkämö PhD, V Leo MA,
M Tervaniemi PhD); CICERO of people with severe age-related brain diseases is stroke, dementia, Parkinson’s disease, epilepsy, or
Learning, University of increasing.1 More than 80% of the heavy economic burden multiple sclerosis. For comparison of the studies
Helsinki, Finland of chronic brain diseases is due to costs other than acute included, we define the effect sizes of improvements in
(M Tervaniemi); Institute of
treatment and care.2,3 This burden has raised the need to the specified outcome as small (Cohen’s d≥0·2), medium
Music Physiology and
Musicians’ Medicine, pursue new cost-effective, easily applicable rehabilitation (d≥0·5), or large (d≥0·8). Effect size was defined as the
University of Music and Drama strategies, both independent of and complementary to mean change in outcome before and after treatment in
Hannover, Hanover, Germany traditional methods such as physiotherapy, occupational the treatment group minus the mean pre–post change
(E Altenmüller MD); and
Division of Clinical
therapy, and speech therapy. Since neurogenesis in the in the control group, divided by the pooled pretest
Neurosciences, Turku adult brain has no known clinically meaningful effect on standard deviation.
University Hospital, Turku, brain recovery, functional restoration relies upon the
Finland (S Soinila) ability of spared neurons to compensate for lost function Music-based interventions for stroke
Correspondence to: by growing neurites and forming new synapses to rebuild Stroke is one of the leading causes of long-term disability
Dr Aleksi J Sihvonen, Cognitive
and remodel the injured networks.4–8 This functional worldwide.23 Of the major neurological disorders, the
Brain Research Unit, Department
of Psychology and Logopedics, restoration is thought to be achieved in traditional strongest evidence for efficacy of music-based interventions
Faculty of Medicine, FI-00014 rehabilitation strategies by targeted training of the has been reported for stroke. We identified 16 randomised
University of Helsinki, Finland weakened function.9–12 An alternative strategy is to increase controlled trials that used music as an add-on therapy
[email protected]
the overall level of brain activity through sensory and for stroke-related neurological and neuropsychiatric
cognitive stimulation.13 disturbances (table).24–39 The assessed outcomes included
Music listening improves neuronal connectivity in motor functions, such as gait and upper extremity
specific brain regions of healthy participants,14–17 and function;24,25,27,30,31,34–39 language functions;26,28,29 cognitive
musical activities, such as playing an instrument, promote functions, such as memory and attention;28,33 mood;28,33,36
neural plasticity and induce grey and white matter changes and quality of life.30,36 These outcomes were measured with
in multiple brain regions, especially front­ otemporal various standard motor tests (eg, Fugl-Meyer Assessment,
areas.18–20 Music listening was efficacious in the recovery of the Box and Blocks Test, Berg Balance Scale, and Nine-
postoperative patients who had various types of major Hole Peg Test), clinical neuropsychological assessments
surgery, as measured by several outcomes such as levels (eg, CogniSpeed, Revised Wechsler Memory Scale),
of pain and anxiety, use of analgesics, and patient standard language function assessments (eg, Boston
satisfaction,21 suggesting that music might also enhance Diagnostic Aphasia Examination), and questionnaires
neurological rehabilitation. (eg, Stroke Impact Scale, Profile of Mood States,
Formal music-based intervention can be defined as and Stroke and Aphasia Quality of Life Scale-39).
active intervention (eg, creating music, playing an Additionally, computer-based movement analyses,30,35,38,39
instrument, singing, or musical improvisation) or MRI analysis,28,31,33 magnetoencephalography,31 or electro­
receptive inter­ vention (eg, music listening) that is encephalography34 were used to assess motor performance
administered by a credentialed music therapist (panel). and neuroplasticity. Metronome-like rhythmic stimulus
Although a Cochrane review evaluated the effect of was used in five studies of stroke-related motor
music intervention on recovery of acquired brain paresis.30,31,36,37,38 The participants’ favourite music, selected
injury,22 a comprehensive overview of music-related through interview, was used in three studies,28,32,33 although
interventions in the rehabilitation of the major the genres of favourite music were not reported.
neurological diseases, including degenerative ones, is Three studies used children’s songs and folk songs.24,34,35
needed. In this Review, we appraise the randomised Five studies involved a trained music therapist.26–28,32,33

648 www.thelancet.com/neurology Vol 16 August 2017

 Review

Effects on motor symptoms

Hemiparesis is the most common consequence of stroke, Panel: Definitions of key terms
affecting over 70% of patients.66 Eight studies reported Music-based intervention
enhanced motor recovery when patients who had a stroke Refers to all experimental protocols that use music in various forms to study its
were rehabilitated with a music-based intervention.25,27,30,34–38 therapeutic effects.
Four of these studies investigated the use of rhythmic
auditory stimulation (panel) in gait training and all four Music therapy
found the intervention to improve gait parameters more Music-based interventions delivered in a clinical setting by a credentialed music therapist
than gait training without any musical support.30,37–39 that use various musical elements (eg, singing, or creating, moving to, or listening to music).
Across studies, significant improvements were observed Music medicine
in gait velocity (d=0·46–2·13), stride length (d=0·49–1·50), Various types of music-based interventions delivered by health-care professionals. The
length of foot contact to surface (d=0·40), cadence interventions have health-promoting goals, but often do not have the therapeutic
(d=0·02–1·82), and symmetry (d=0·52–0·83) after relationship and reciprocal musical interaction that are characteristic of music therapy.
3–6 weeks of rhythmic auditory stimulation compared
with conventional training without rhythmic auditory Rhythmic auditory stimulation
stimulation (table).37–39 There were similar findings when Neurological music therapy technique used in the rehabilitation of movements that are
intensive gait training with rhythmic auditory stimulation naturally rhythmic (eg, gait). In rhythmic auditory stimulation, a series of auditory stimuli
was investigated with respect to postural control and gait are presented at a fixed rhythm, and movements are synchronised (entrained) to the
performance in patients with chronic stroke.30 In one rhythm.
study,30 over the course of 6 weeks of intervention, the Music-supported therapy
rhythmic auditory stimulation group improved in Music-based intervention developed for motor rehabilitation of stroke. In music-supported
balance, gait velocity, cadence, stride length, and double- therapy, gross and fine movements of the hemiparetic upper extremity are trained through
support period on the affected side. In another study,36 playing musical instruments (eg, drums, keyboard).
when rhythmic auditory stimulation was used in the
form of combining rhythmic music with movement Melodic intonation therapy
therapy, patients who had a stroke showed improved Singing-based intervention developed for rehabilitation of people with non-fluent aphasia.
ankle (d=0·61) and arm movement (d=0·99) after 8 weeks In melodic intonation therapy, the intonation of speech is expressed as low and high
of intervention. One study compared bilateral arm pitches. Production of linguistic phrases is achieved by training with a music therapist, first
training with rhythmic auditory stimulation to dose- by intoning (singing) them at a slow pace and steady rhythm supported by tapping, then
matched therapeutic exercises, but there were no following a hierarchy of steps that eventually transforms communication from singing to
significant differences between the groups.31 Numerically, speech. Training starts with singing two-syllable words and proceeds gradually to phrases.
rhythmic auditory stimulation done by a music therapist
(compared to control) resulted in greater improvement
compared with stimulation by a non-music therapist feedback therapy.24 Although both rhythmic auditory
(compared to control).22 stimulation and music-supported therapy involve
Music-supported therapy (panel) was efficacious in auditory-motor coupling, incorporating music stimulus
rehabilitating arm paresis after stroke in five randomised might result in enhancement of the therapeutic effect
controlled trials.24,25,27,34,35 3 weeks of music-supported because of the personal motivational value of music.
therapy improved motor skills in the patient’s paretic arm Internal synchronisation, based on musical memory,
significantly more than conventional physiotherapy, generates the expectation of consecutive sounds of a
an effect shown by several validated clinical tests familiar song, and provides precise mental timing
(Cohen’s d=0·24–0·69).34,35 The effect was accompanied feedback for movement, thus supporting the patient’s
by improved cortical connectivity and increased activation impaired proprioception.
of the motor cortex.34 The improvement in motor skills
observed seemed to be specifically caused by music rather Effects on aphasia
than by motor training, since patients practicing with Aphasia affects approximately 30% of patients who have
mute instruments showed less improvement than the had a stroke.66 In two randomised controlled trials, active
music group.27 music therapy improved the speech of people with
One study25 used movement sonification therapy, a chronic aphasia.26,29 In one study, melodic intonation
recent development in music-supported therapy in which therapy (panel),67 a singing-based speech therapy
gross movement is transformed into sound, providing designed for non-fluent people with aphasia, was given
continuous feedback, substituting for deficits in proprio­ to 10 of 17 people with subacute aphasia.29 In this study,
ception. Movement sonification therapy reduced joint pain melodic intonation therapy improved the patient’s daily
(Cohen’s d=1·96) and improved smoothness of movement life communication (Cohen’s d=0·76) and object naming
(d=1·16) more than movement therapy without sound. (d=1·73) significantly more than those outcomes in the
Delayed auditory feedback in music-supported therapy control group, which received other types of language
could be as effective as traditional immediate auditory rehabilitation.29 Music-related speech therapy, particularly

www.thelancet.com/neurology Vol 16 August 2017 649

 Review

melodic intonation therapy, is conceptually elegant and the help of a music therapist, and continued during the
music therapy interventions might be more effective in first 2 months after stroke, enhanced cognitive recovery.33
aphasia than speech training without music.22 In a 6-month follow-up, the music group still showed
significant improvements in performance of tasks
Effects on cognitive and emotional deficits measuring verbal memory (Cohen’s d=0·88) and focused
Deficits in cognitive functions (eg, memory, attention, attention (d=0·92), compared with a control intervention
executive function) and mood (eg, depression) affect (audiobook listening) or standard care.33 Compared with
approximately 30–50% of patients who have experienced standard care, music listening was also associated with
a stroke.68,69 In one randomised controlled trial, 1 h daily less depression (d=0·77) and confusion (d=0·72;
listening to a participant’s favourite music, selected with figure 1A).33 The cognitive gains measured after music

Studies Partici­ Music Blinding Primary outcome Overall Main results

(n) pants therapist duration of
(n) involved intervention
Stroke
MST vs MST with delayed 1 34 No Single Hand movement 5 h in 4 weeks No significant difference between the groups.
sound24
Sonificated movement vs 1 25 No No Gross motor function 10 days Sonification therapy reduced joint pain (p<0·05, d=1·96) and
movement without sound25 improved movement smoothness (p=0·04, d=1·16).
Singing, playing 1 20 Yes No Speech parameters in people 22·5–37·5 h in Music therapy improved spontaneous speech (p=0·020, d=0·35).
instruments, improvisation with chronic aphasia 15 weeks
vs speech therapy26
MST vs playing mute 1 33 Yes No Upper-limb motor function 20 sessions in MST improved motor functions (p=0·039).
instruments 27 4 weeks
Favourite music vs 1 49 Yes Single Grey matter changes 60 h in Music listening increased grey-matter volume in frontal areas, limbic
standard care28 associated with cognitive 8 weeks areas, and right ventral striatum. Reorganisation in the frontal areas
improvement correlated with enhanced recovery of verbal memory, focused attention,
and language skills, whereas the limbic area reorganisation correlated
with reduced negative mood.
Melodic intonation 1 27 No Single Speech parameters in 30 h in Melodic intonation therapy improved daily life communication
therapy vs other language non-fluent aphasics 6 weeks (d=0·76) and object naming (d=1·73).
intervention29
RAS vs intensive gait 1 20 ·· No Postural control and gait 15 h in RAS improved balance, gait velocity, cadence, stride length, and
training30 performance 6 weeks double support period on the affected side, and improved score on
stroke-specific quality-of-life scale.
Bilateral arm training with 1 92 No Single Functional reorganisation 18 h in There were no significant differences between the groups.
rhythmic auditory cueing and outcome 6 weeks
vs normal exercise31
Favourite music and 2 54 Yes Single Auditory sensory memory;32 60 h in Listening to music and speech after neural damage can induce
audiobook listening vs cognitive functions and 8 weeks long-term plastic changes in early sensory processing;32 music listening
standard care32,33 mood33 improved verbal memory (p=0·002, d=0·88) and focused attention
(p=0·012, d=0·92) compared with the audiobook and control groups.
Music group also had less depression (p=0·031, d=0·77) and confusion
(p=0·045, d=0·72) than the control group.33
Children’s music and folk 2 62;34 No No Neuroplasticity and motor 7·5 h in MST improved motor skills showed by action research arm test score
songs vs conventional 4035 recovery;34 motor recovery35 3 weeks (p<0·001, d=0·32), arm paresis score (p<0·05, d=0·46), Box and Blocks
therapy34,35 test (p<0·001, d=0·43), and Nine-Hole Peg Test (p<0·05, d=0·32);34 music
group improved in speed, precision, smoothness of movements, and
motor control in everyday activities evaluated by action research arm test
(p<0·001, d=0·36), arm paresis score (p<0·05, d=0·42), Box and Blocks
test (p<0·001, d=0·69), and Nine-Hole Peg test (p<0·05, d=0·24).35
RAS vs referral 1 33 No No Upper and lower limb 16 h in RAS improved range of ankle extension (p=0·018, d=0·61) and arm
information36 mobility, mood, 8 weeks flexibility up (p=0·001, d=0·99) and down (p=0·008, d=0·62), mood
interpersonal relationships, (p=0·017, d=0·03), and increased frequency and quality of
and quality of life interpersonal relationships (p=0·003, d=0·96).
RAS vs neurodevelopmental 1 78 No Single Gait parameters 7·5 h in RAS improved velocity (p=0·006, d=2·13), stride length (p<0·001,
therapy37 3 weeks d=1·50), cadence (p<0·001, d=1·82), and symmetry (p=0·049, d=0·83).
RAS vs gait training without 1 23 No No Gait parameters 5 h in 3 weeks RAS improved gait velocity (p=0·008, d=0·46), stride length (p=0·009,
musical feedback38 d=0·49), cadence (p=0·045, d=0·02), symmetry (p=0·008, d=0·55),
and heel–toe distance (p=0·006, d=0·40).
RAS vs physical therapy39 1 20 No Single Gait parameters 30 h in RAS improved gait velocity (d=1·45), stride length (d=0·93), symmetry
6 weeks (d=0·52), and cadence (d=0·44).
(Table continues on next page)

650 www.thelancet.com/neurology Vol 16 August 2017

 Review

listening were associated with enhanced auditory had not been affected by the stroke28 (figure 1B, 1C).
memory-related function in temporal brain areas32 and Reduction in negative mood after music listening was
increased grey matter volume in prefrontal regions that associated with increased grey matter volume in limbic

Studies Partici­ Music Blinding Primary outcome Overall Main results

(n) pants therapist duration of
(n) involved intervention
(Continued from previous page)
Dementia
Multisensory stimulation 1 18 No No Neuropsychiatric symptoms 16 h in Multisensory stimulation showed positive effects on anxiety
vs music listening40 and cognition 16 weeks symptoms and dementia severity that were not observed in the
music listening group.
Music listening vs singing 3 83 Yes Single Emotional parameters;41 15 h in Both music listening and singing groups improved in behavioural
vs standard care41–43 clinical, demographic, and 10 weeks disturbances (p=0·04, d=0·42) and physical signs (p=0·008, d=0·52)
musical background factors more than the control group. Effects not present 6 months after the
influencing the cognitive intervention;41 singing was beneficial, especially in improving working
and emotional efficacy of memory in people with mild dementia and in maintaining executive
caregiver-implemented function and orientation in young people with dementia. Music
musical activities;42 listening was beneficial in supporting general cognition, working
quality of life, mood, and memory, and quality of life, especially in people with moderate
cognition43 dementia not caused by Alzheimer’s disease who were in institutional
care. Both music interventions alleviated depression, especially in
people with mild dementia and Alzheimer’s disease. The musical
background of people with dementia did not influence the efficacy of
the music interventions;42 music listening improved the patients’
mood (p=0·001, d=0·80), orientation (p=0·005, d=0·71), episodic
memory (p=0·036, d=0·54), attention and executive functions
(p=0·039, d=0·48), overall cognitive performance (p=0·041, d=0·47),
and the quality of life (p<0·001, d=0·99). Singing resulted in additional
improvement in short-term memory and working memory (p=0·006,
d=0·75), and improved caregiver wellbeing (p=0·026, d=0·85).43
Music therapy and music 1 98 Yes Single Behavioural and 10 h in No significant differences between the groups.
listening vs standard care44 psychological symptoms of 10 weeks
dementia
Music listening, singing, 1 13 Yes No Neuropsychiatric symptoms, 11 h in Music group showed improvement in symptoms (p=0·002, d=2·32)
improvising, and talking vs well-being, and 22 weeks and in levels of wellbeing (p<0·001, d=3·85). Staff in the intervention
standard care45 carer–resident interaction group reported enhanced caregiving techniques as a result of the
programme.
Group music therapy vs 1 100 Yes Single Mood and cognition 6 h in 6 weeks Group music therapy decreased depression (p=0·001, d=0·21) and
standard care46 delayed the deterioration of cognitive functions, especially recall
(p=0·004, d=0·72). The effects were present 1 month after cessation of
the intervention.
Music therapy (listening 3 76;47 Yes Single Neuropsychiatric 21 h in Neuropsychiatric symptoms decreased significantly in the music therapy
and singing) vs other 77;48 symptoms;47 agitation;48 16 weeks;47,48 group (p=0·01);47 there were no significant differences between the
activities47–49 5949 behavioural and 15 h in groups;48 music therapy improved behavioural symptoms (p<0·0001,
psychological symptoms49 16 weeks49 d=1·04), functional ability (p<0·0001, d=0·79), and empathetic
behaviour (p<0·0001, d=0·61) compared with the control treatment.49
Music therapy (listening, 1 37 ·· Single Patients’ mood, cognition, 8 h in 4 weeks There were no significant differences between the groups.
playing and singing) vs behavioural disturbances,
cooking50 and stress experienced by
their nurses
Music therapy vs standard 2 50;51 Yes Single Cognition and anxiety; 51 18 h in The music group improved performance in attention (p=0·001, d=0·76)
care51,52 5052 behavioural disturbances52 12 weeks;51 and verbal episodic memory tasks (immediate p=0·001, d=0·76; delayed
6 h in 4 weeks52 p=0·001, d=0·73), but not in anxiety;51 music reduced the behavioural
disturbances showed by significant group difference (p<0·05, d=0·63).52
Favourite music vs 1 52 No No Anxiety 6 h in 6 weeks Anxiety decreased in the music group (p=0·004, d=0·06).
standard care53
Music therapy (playing and 1 100 Yes No Agitation 6 h in 6 weeks There was no significant differences between the groups.
listening) vs standard care54
Music therapy (listening 1 47 Yes Single Mood and quality of life 32 h in There were no significant differences between the groups.
and playing) vs reading55 16 weeks
Music therapy vs resting 1 30 Yes Single Anxiety and mood 5 h in Music therapy decreased anxiety (p<0·001, d=2·42) and depression
and reading56 16 weeks (p=0·002, d=1·05). These effects persisted up to 2 months after
stopping the intervention.
(Table continues on next page)

www.thelancet.com/neurology Vol 16 August 2017 651

 Review

Studies Partici­ Music Blinding Primary outcome Overall Main results

(n) pants therapist duration of
(n) involved intervention
(Continued from previous page)
Parkinson’s disease
Music listening, rhythmic 1 18 Yes Single Motor performance, 12 h in Music therapy improved mobility (p=0·006), UPDRS III (p=0·003), text
clapping, or stomping vs cognition, and quality of life 6 weeks recall (p=0·036), item naming (p=0·033), performance in Stroop test
standard care57 (p=0·007), and quality of life (p=0·031).
Favourite music 1 22 No Single Walking parameters 19·5 h in Walking to music improved velocity (p=0·002, d=2·64), stride time
synchronised to gait vs 13 weeks (p=0·019, d=1·76), cadence (p=0·007, d=2·16), and UPDRS III (p=0·002,
regular activities58 d=0·50).
Tango, waltz, or foxtrot 1 48 No Single Functional motor control 20 h in Tango group improved in balance (p=0·001, d=2·98), 6-min walking
dancing vs standard care59 13 weeks (p=0·001, d=2·50), and backward stride length (p=0·001, d=2·19); waltz
or foxtrot group improved in balance (p=0·001, d=3·17), 6-min walking
(p=0·001, d=2·24), and backward stride length (p=0·018, d=1·96).
Tango and waltz or foxtrot 1 61 No No Health-related quality of life 20 h in Tango improved mobility (p=0·03, d=2·50), social support (p=0·05,
vs Tai Chi or standard care60 13 weeks d=2·97), and quality of life (p<0·01, d=2·09).
Tango vs physical exercise61 1 19 No Single Functional mobility 20 h in Tango group improved balance (p=0·01, d=2·18).
13 weeks
Multiple sclerosis
Keyboard playing vs mute 1 19 No No Hand function 7·5 h in The music group improved in the functional use of the hand significantly
keyboard playing62 2 weeks more showed by time × group interaction (p=0·003, d=0·60).
RAS vs standard care63 1 10 Yes No Gait parameters 2 weeks RAS significantly decreased double-support time (left: p=0·018,
d=1·61; right: p=0·025, d=1·46).
Epilepsy
Nightly exposure of 1 73 No Single Seizure occurrence Every night for Seizure frequency in the music group decreased significantly during
Mozart Sonata K. 448 vs 1 year the treatment phase (17%, p=0·014) and 1 year after treatment
no intervention64 (16%, p=0·027).

Effect size (Cohen’s d) was the mean change in outcome before and after treatment in the treatment group minus the mean pre–post change in the control group, divided by the pooled pre-test SD.65 Effect size was
defined as small (d≥0·2), medium (d≥0·5), and large (d≥0·8). d=Cohen’s d. MST=music-supported therapy. RAS=rhythmic auditory stimulation. UPDRS III=Unified Parkinson’s Disease Rating Scale part III.

Table: Randomised controlled trials assessing various music-based interventions in patients with stroke, dementia, Parkinson’s disease, multiple sclerosis, or epilepsy

areas.28 In addition to music listening, rhythmic auditory Impairment Battery), and quality of life or wellbeing
stimulation therapy improved patients’ mood but the (Cornell–Brown Scale for Quality of Life in Dementia,
effect size was not significant.36 Although positive effects Dementia Care Mapping). Most interventions used vocal
were shown to last up to at least 6 months for some or instrumental music that was presumably familiar to
outcome measures, these results need to be replicated. the participants, such as personal favourites, general
popular music, or common children’s songs. In all
Music-based interventions for dementia studies except for one, the intervention was administered
The most common causes of dementia are Alzheimer’s by a music therapist or music teacher.
disease, cerebrovascular diseases, or a combination of
the two. In these disorders, neural degeneration can Effects on cognitive deficits
progress over several years, leading to memory problems In four studies, music listening coupled with cognitive
and other behavioural disturbances. So far, 17 randomised elements (reminiscence and attention training) or
controlled trials enrolling people with dementia (table) physical exercise improved overall cognitive performance
have assessed the effects of music-based interventions (measured by MMSE) of patients with dementia,
on neuro­ psychiatric and behavioural symptoms such compared with standard care (Cohen’s d=0·47–0·76).42,43,46,51
as anxiety and agitation (14 studies),40–42,44,45,47–54,56 depression Add­ itionally, improved performance in these music
(six studies),42,43,46,50,55,56 cognitive status (five studies),42,43,46,50,51 interventions was reported for tests measuring
and quality of life (four studies).42,43,45,55 Neuropsychiatric attention and executive functions (d=0·48–0·76),43,51
and behavioural symptoms were assessed with tests, orientation (d=0·71),43 and verbal or episodic memory
rating scales, or questionnaires measuring overall (d=0·54–0·76).43,51 In one randomised controlled
symptom severity (eg, Neuropsychiatric Inventory, trial,43 caregiver-implemented singing enhanced short-
Cohen–Mansfield Agitation Inventory, Behavioural term and working memory (d=0·75), especially in
Pathology in Alzheimer’s Disease rating scale), participants with mild dementia. It also reduced
depression (eg, Cornell Scale for Depression in caregiver burden (assessed with Zarit Burden Interview;
Dementia, Geriatric Depression Scale), cognitive status d=0·85). By contrast, no significant changes in cognitive
(eg, Mini-Mental State Examination [MMSE], Severe performance were observed for group-based music and

652 www.thelancet.com/neurology Vol 16 August 2017

 Review

A B
Verbal memory (correct responses) Focused attention (correct responses) Baseline
35
20 1 nAm
30 p<0·01 p<0·05

15
Change score

25
Change score

20 10
15 3 months
5
10
0
5

0 –5
Baseline 3 months 6 months Baseline 3 months 6 months
6 months

Depression Confusion
10 8

8
6
p<0·005 Baseline
6
6 3 months
Score

p<0·01 6 months
Score

4
MMN amplitude (nAm)

4
4
2 p<0·1
2 p<0·05
p<0·1
p<0·05
2
0 0
3 months 6 months 3 months 6 months

Music group 0
Audiobook group Music group Audiobook Control
Control group group group

C
Left ventral/subgenual anterior congulate Right superior frontal gyrus
0·03
0·02
GMV 6m-acute

0·01
0
–0·01
–0·02
–0·03
Left hemisphere x=–12 Left hemisphere z=50
Left superior frontal gyrus Right medial superior frontal gyrus
0·03
0·02
GMV 6m-acute

0·01
0
–0·01
–0·02
–0·03 Left hemisphere y=23 Left hemisphere z=38
Music Audiobook Control Music Audiobook Control
group group group 0 3 6 T-value group group group

Number of patients 2 3 4

Figure 1: Effects of daily music listening after stroke

Cognitive, emotional, and neural effects of daily music listening (music group), audiobook listening (audiobook group), and standard care (control group) 1 week (BL),
3 months, and 6 months after stroke. (A) Neuropsychological results (mean and SEM) showing improved recovery of verbal memory and focused attention (baseline
score subtracted from the values) and less depression and confusion in the music group compared with the audiobook and control groups. p values for verbal memory
and focused attention calculated with mixed-model ANOVA. p values for depression and confusion calculated with one-way ANOVA. Adapted from Särkämö and
colleagues.33 (B) Magnetoencephalography group results (mean ± SEM) showing increased right hemisphere MMN responses to pitch changes in the music and
audiobook groups compared with the control group. Adapted from Särkämö and colleagues.32 (C) Voxel-based morphometry results of MRI data from patients with
left-hemisphere damage (lesion areas in blue–green), showing larger GMV increases (mean ± SEM) in prefrontal and limbic areas in the music group compared with the
audiobook and control groups. Results are shown at p<0·01 (uncorrected) with 50 voxels or more of spatial extent. T value shows GMV increases for the music group
compared with the audiobook and control groups (Group × Time interaction). Adapted from Särkämö and colleagues.28
BL=baseline. nAm=nanoamperes. MMN=mismatch negativity. GMV=grey- matter volume.

www.thelancet.com/neurology Vol 16 August 2017 653

 Review

cooking interventions in people with moderate-to-severe dementia-level cognitive decline in the late stage of the
dementia.50 The cognitive benefits of music observed disease.71 The effect of music on several symptoms of
only in the early stages of dementia might be related Parkinson’s disease have been studied in five randomised
to enhanced cognitive reserve, the use of alternative controlled trials (table).57–61 Four studies examined the
networks, and cognitive strategies to cope with advancing effects of music-assisted motor training using motor
pathology.70 parameters as outcome measures.57–59,61 Two studies57,60
evaluated non-motor parameters, quality of life,
Effects on neuropsychiatric symptoms, mood, and cognition, or social parameters. In all studies, medication
quality of life for Parkinson’s disease remained unchanged during the
Results from six studies showed that music therapy was interventions.
efficacious in improving the neuropsychiatric symptoms General motor performance was assessed using the
of people with dementia (Cohen’s d=0·42–2·32).41,42,45,47,49,52 motor part of the Unified Parkinson’s Disease Rating
Three of these studies assessed the duration of effect Scale (UPDRS-III), and specific motor functions using,
after cessation of the intervention,41,46,56 which varied for example, the Berg Balance Scale and 6-min walk test.
from less than 4 weeks to 2 months. In one of the Specific gait parameters were analysed using video
six studies, the music intervention programme also recordings and computer-assisted motion analysis
resulted in improved participant–caregiver interaction, programs. Quality of life was evaluated using validated
measured by semi-structured interview, and improved questionnaires. Music used in the intervention varied
participant wellbeing (d=3·85).45 By contrast, two studies from rhythmic auditory cueing to self-selected favourite
did not show any significant effect of music therapy or music, although the genres of the patient’s favourite
music listening on neuropsychiatric symptoms.44,50 music were not reported. A music therapist administered
Regarding specific neuropsychiatric symptoms, results the intervention in only one study.57
from two studies showed that music reduced anxiety Based on effect sizes we calculated from the published
and agitation in people with dementia,53,56 but the effect data, dancing had the most consistent and clinically
sizes diverged (d=0·06 vs d=2·42). By contrast, results significant beneficial effect on motor symptoms. In one
from four randomised controlled trials showed music to study of 48 participants,59 both tango and waltz or foxtrot
be ineffective in reducing anxiety or agitation.40,48,51,54 intervention groups improved in balance (Cohen’s
Quality of life was assessed in three studies.42,43,55 d=2·98 for tango, d=3·17 for waltz or foxtrot), 6-min
Although Cooke and colleagues55 did not find any walk test (d=2·50, d=2·24), and backward stride length
significant differences between the effects of music and (d=2·19, d=1·96) compared with standard care.59 In a
the control intervention (interactive reading), Särkämö smaller study of 19 participants,61 tango improved
and colleagues43 reported that music listening compared balance (d=2·18). Dancing also improved overall
with standard care substantially increased quality of life mobility in another study (d=2·50).60 Music therapy with
(Cohen’s d=0·99),43 especially in people with moderate rhythmic movements—which uses a coupling of
dementia with causes other than Alzheimer’s disease.42 movement and music that is similar to dancing—
Decreased depression or improvement of mood in improved overall mobility in patients with Parkinson’s
people with dementia has been reported in four studies disease.57 In another study,58 gait training synchronised
(d=0·21–1·05).42,43,46,56 Two other randomised controlled to music resulted in improved velocity (d=2·64), stride
trials did not show such an effect.50,55 In both of these time (d=1·76), and cadence (d=2·16).58 Both studies
studies, the control group received an intervention other reported a reduction in Parkinson’s disease-specific
than standard care, raising the possibility that any motor symptoms (d=0·50).57,58
pleasurable activity improved mood. Overall, the effects Results from two studies57,60 showed that a music-based
of musical interventions in dementia could be driven by intervention improved quality of life, with large effect
the comfort and emotional safety induced by familiar size. Dancing the tango appeared to be significantly
music, which can temporarily overcome confusion and more effective than waltz or foxtrot, tai chi, or regular
disorientation by anchoring a person’s attention on a treatment (d=2·09).60 Additionally, patients reported
positive familiar stimulus in an otherwise confusing better social support after the intervention (d=2·97).
environment. We speculate that this anchoring effect Improvements in cognition were reported in one study.57
could be enhanced by using headphones. Familiar music Although the sample sizes in these studies were
can also be imbued with emotions that are specific to an relatively small, the evidence suggests that dancing and
individual, and can trigger autobiographical memories music-based interventions that synchronise movement to
and help to temporarily restore a sense of identity. music can be beneficial in the maintenance of motor
performance in people with Parkinson’s disease.
Music-based interventions for Parkinson’s disease Rhythmical use of musical stimuli compensates for the
Patients in the early stage of Parkinson’s disease can loss of control by the extrapyramidal system and enhances
have deficits of the autonomic nervous system and audio perception and movement synchronisation.30,37,38
other non-motor deficits, and 30% of patients develop The perceived rhythm in music activates the neural

654 www.thelancet.com/neurology Vol 16 August 2017

 Review

Somatosensory cortex Motor cortex Insula

Thalamus
Inferior parietal lobe Premotor cortex Precuneus
Cingulate gyrus
Angular gyrus
Dorsolateral
prefrontal cortex
Medial
prefronal
cortex

Planum temporale
Hippocampus
Orbitofrontal cortex
Inferior frontal gyrus
Cerebellum Striatum
Superior temporal gyrus
Ventral tegmental area Nucleus accumbens
Middle temporal gyrus
Amygdala
Auditory cortex Inferior colliculus

Basic auditory pathway: perceiving the basic acoustic features of music

Musical–syntactic network: perceiving higher-order musical features
Attention and working memory network: focusing and keeping track of music in time
Episodic memory network: recognising music and recalling associated memories
Motor network: playing, singing, and moving to the beat of music
Reward and emotion network: music-evoked emotions and experiencing pleasure and reward

Figure 2: Key brain areas associated with music processing

Areas identified from neuroimaging studies of healthy people. Although the figure displays the lateral and medial parts of the right hemisphere, many musical
processes are largely bilateral (with the exception of pitch and melody processing, which are lateralised, the activity in the right hemisphere being dominant).
Adapted from Särkämö and colleagues.16

circuits involved in motor actions and acts as an external The trial without a music therapist62 included
cue for movement, thus replacing the impaired internal 19 patients and studied the effect of keyboard playing
timing function in people with Parkinson’s disease.72 As (audible vs mute) in hand functionality. Audible keyboard
in studies of stroke rehabilitation,22 the use of music as a playing significantly improved the functional use of the
stimulus might be more effective than auditory hand (Cohen’s d=0·60), as indicated by a validated
stimulation without music (eg, metronome beat) in gait questionnaire (ABILHAND). Using a computerised gait
rehabilitation, but requires further investigation. analysis, a feasibility study63 of ten patients with multiple
Similarly, external cueing might also explain the positive sclerosis and gait problems found rhythmic auditory
effects of dancing in people with Parkinson’s disease. stimulation to be effective in decreasing double-support
Furthermore, the improvement in motor control and time (d=1·46–1·61).63 Although decreased double-support
possible decrease in disease-specific symptoms might in time might reflect improved dynamic balance,73 none of
turn improve quality of life. In all reviewed studies, the the other gait parameters tested in this study differed
follow-up period was too short to draw conclusions on the from controls receiving standard care. The results of
long-term effects of music interventions. The effects of music-based interventions in multiple sclerosis are scant
music on the autonomic disturbances in Parkinson’s and allow no definite conclusions on the rehabilitative
disease have not been addressed in controlled studies. effect of music. Although designing studies can be
challenging because of the diversity of disease deficits,
Music-based interventions for multiple sclerosis motor functions, spasticity, fatigue, cognitive deficits,
Multiple sclerosis is one of the most common neurological and mood might be feasible outcome measures in the
diseases in the young adult population, with disease onset future studies.
occurring in most cases between the ages of 20 and
40 years. Despite relatively low prevalence, patients Music-based interventions for epilepsy
with the disease require expensive medication and in Epileptic seizures arise from abnormal synchronisation
most cases life-long rehabilitation.3 Multiple sclerosis of electrical activity in the brain, and most of them
treatments aim to ameliorate function after flare-up of an cease spontaneously, by largely unknown mechanisms.
episode or to prevent new episodes. Only two randomised Exposure to patterned auditory stimuli provides a non-
controlled trials62,63 have studied the effect of musical invasive excitatory stimulation of the cortex, which might
interventions in alleviating the manifestations of multiple reduce epileptiform activity.74 To test this hypothesis, one
sclerosis (table). Between these studies, outcomes were randomised controlled trial (table) examined the efficacy
different, and the intervention was administered by a of music in patients with epilepsy.64 Patients were exposed
music therapist in only one study. to Mozart’s music at periodic intervals every night for

www.thelancet.com/neurology Vol 16 August 2017 655

 Review

(figure 3), which should provide favourable circumstances

Increased activity
for recovery in general. For example, after stroke,
neuroplastic changes associated with functional recovery
are activity-dependent.77 Musical activities bear similarity
Increased cerebral
blood flow
with the concept of enriched environments used in
animal studies, which facilitates recovery at behavioural
Striatum and neurobiological levels in animal models of many
neurological illnesses.13
Nucleus accumbens Since active music-based rehabilitation involves
Hippocampus multiple components analogous to training and music
Hypothalamus learning (ie, iterated practice of movements coupled with
auditory feedback and extensive cognitive processing), it is
Ventral tegmental area Pituitary gland plausible that music-based neurological rehabilitation
Amygdala
induces similar structural and functional neuroplastic
changes seen in populations of healthy individuals that
Hypothalamus receive musical training.18,19 Indeed, some studies have
HPA-axis
reported memory-related plastic effects after music
Brainstem Hypothalamus listening,28,32 as well as neural reorganisation after music-
CRH supported therapy in patients recovering after a stroke.34
Autonomic nervous system Pituitary gland Other studies have provided further evidence of auditory-
ACTH related and motor-related neuroplasticity after music-
Cardiovascular responses Adrenal cortex supported therapy78–80 and melodic intonation therapy81 in
patients who have had a stroke.
Cortisol However, the specific cellular mechanisms of music-
induced neuroplasticity remain unknown. Although
Figure 3: Possible neurobiological mechanisms for the rehabilitative effect of music
substantial neurogenesis in elderly individuals seems
Orange circles and yellow arrows represent the mesolimbic system, and the green circles represent the HPA axis. unlikely, other putative mechanisms include neuronal
ACTH=adrenocorticotropic hormone. CRH=corticotropin-releasing hormone. hypertrophy, increased volume of neuropil, and changes
HPA axis=hypothalamic-pituitary-adrenal axis. in the vascular or glial compartments. An intriguing
question to investigate would be whether previous music
1 year, and a significant 17% reduction in seizure exposure during a specific period of lifetime affects the
frequency was detected during the study period plasticity of the recovering brain. The possibility of
(p=0·014). Additionally, a carry-over effect of 16% reduced negative plastic changes due to overly intense or
seizure frequency persisted for 1 year. Although no other premature intervention should be considered.
randomised controlled trials with adult participants have
been published, a recent meta-analysis of 12 studies Activation of reward, arousal, and emotion networks
including both children and adult patients with epilepsy Music activates the dopaminergic mesolimbic system,
of any kind indicated that 130 (85%) of 153 patients which regulates memory, attention, executive functions,
responded favourably to music, with an average reduction mood, and motivation82 (figure 3). A key part of this
in interictal epileptic activity of 31% during the listening reward system is the nucleus accumbens, which
period and 24% after the listening period.75 Further regulates mood and pleasure. In healthy individuals, its
studies are needed, since only two of these studies had a activation by an intense emotional response to music (so-
separate control group. called chills) leads to increased dopamine secretion
directly proportional to the intensity of the experience.82
Mechanisms underlying the rehabilitative effect Increased extracellular dopamine levels could partly
of music explain the cognitive–emotional gains induced by music
Considering the widely varying nature of the diseases in in patients with neurological disorders. Music-induced
which music has led to improved recovery, enhanced improvement of mood, arousal, and relief of confusion
rehabilitation, or alleviation of symptoms, several distinct might therefore enhance recovery of cognitive functions
explanatory mechanisms can be postulated. in these patients. Music-induced activation of the
parasympathetic nervous system and inhibition of the
Neural activation and neuroplasticity sympathetic nervous system in people with dementia,
Results from functional neuroimaging studies in healthy and corresponding changes in catecholamine and
participants have shown that music induces widespread cytokine secretion, has been considered as a soothing
activation of various networks in the brain14–17 (figure 2), effect of music.83 This increased parasympathetic activity
and correspondingly increases blood flow through the is also a possible mechanism behind the effect of music
medial cerebral artery because of autoregulation76 ameliorating neuropsychiatric symptoms in dementia.

656 www.thelancet.com/neurology Vol 16 August 2017

 Review

Music also produces measurable cardiovascular and Familiar music specifically activates the anterior
endocrine responses, indicated by reduced serum cingulate and medial prefrontal cortex in the healthy
cortisol levels and inhibition of cardiovascular stress brain, suggesting that these regions are important in
reactions83,84 (figure 3). In animal models, prolonged musical memory.95 In people with Alzheimer’s disease,
stress can have maladaptive effects on neuroplasticity, the medial prefrontal cortex degenerates more slowly
such as dendritic atrophy, synapse loss, and decreased than do other cortical regions, and the regions that
hippocampal neurogenesis.85 Elevated cortisol levels in encode musical memory also show minimal atrophy or
patients with acute stroke correlates with increased decrease in glucose metabolism, despite amyloid-β
infarct volume, and increases the risk of depression, deposition.95 These observations provide a potential
poor prognosis, and fatal outcome.86 We speculate that explanation for why patients with Alzheimer’s disease
listening to music might lower stress hormone secretion are able to recognise and respond emotionally to familiar
in acute stroke, as it does in postoperative patients.87,88 songs, even at late stages of the disease.95
Overall, neurological diseases and mood disorders
have a high comorbidity, ranging from 20% to 50%.89,90 Conclusions and future directions
Common clinical experience is that depression Long-term treatment and rehabilitation for patients with
diminishes adherence to rehabilitation, and studies neurological diseases accounts for a substantial proportion
indicate that depression impairs functional outcome and of the associated costs, and therefore, study of novel
quality of life, and increases mortality.91 According to the rehabilitation strategies to replace or complement
data discussed previously, music improved mood or traditional methods is warranted. With this aim, the effects
diminished anxiety in people with dementia43,53 and in of music-based rehabilitation in major neurological
patients who have had a stroke.33,36 We conclude that disorders have been studied in 41 randomised controlled
music interventions are viable in improving the mood of trials to date. Music interventions seem to be beneficial,
patients with neurological disorders. However, the causal particularly in motor rehabilitation for people with stroke
relationship between music-induced mood improvement and Parkinson’s disease. Additionally, music interventions
and neurological outcome remains to be proved. can have favourable effects on cognition, mood, and
quality of life in people with stroke or dementia.
Activation of alternative or spared neural networks Although the majority of studies have reported positive
Some music interventions engage specific regions effects, the possibility of publication bias should be
associated with musical rhythm, movement, singing, or considered. In addition, only a few of the primary
memory that are not directly affected by the disease.72 outcomes have been studied repeatedly. Limitations
Rhythmic entrainment, our inherent tendency to time in most studies arise from small sample sizes and
movements to the regular beat of music, which forms methodological heterogeneity in study design, and in the
the basis of rhythmic auditory stimulation and playing- interventions and outcome measures used. In most
based music interventions, is based on the strong studies, the duration of the music-induced rehabilitation
connectivity between the auditory system and motor effect was not systematically evaluated and is still largely
system.14 In diseases in which the internal sequencing unknown. Thus far, music-based interventions have
and monitoring of actions does not work because of been observed to have long-term effects in stroke
motor system dysfunction, rhythmic entrainment can (3 months),33 dementia (maximum 2 months),41,46,56 and
act as an external timer, cueing the execution of epilepsy (12 months).64
movements, and therefore bypassing the dysfunction.72 In some studies, the difference between active and
For instance, a patient with impaired muscle receptive intervention and the role of the music therapist
coordination after a stroke, or a patient with Parkinson’s (if participating) remained unclear. The therapeutic
disease with stiffness and bradykinesia, might find it relationship between patient and therapist that is inherent
easier to execute motor tasks with rhythmic support in formal music therapy is likely to have an additional
provided by music listening or dancing.30,37–39,57–59,61 effect on the outcome. Although this aspect is difficult to
Singing, which is the key component of melodic delineate from the music intervention used, the outcome
intonation therapy, engages frontotemporal language of an intervention given by a music therapist might in
and vocal-motor regions bilaterally, and more extensively, some cases be superior to that given by another health-
than does speaking.92,93 This engagement enables training care professional, as has been observed for rhythmic
of speech in patients with aphasia via both spared left auditory stimulation in gait rehabilitation.22 However, the
hemisphere regions and homologous right hemisphere studies reviewed here showed that both music therapy
regions. The preserved ability to sing in patients with and other music-based interventions have beneficial
aphasia has been reported as early as in 1745, when a effects regardless of the involvement of a dedicated music
patient with severe aphasia after a stroke was reported to therapist. Most of the studies did not have an adequate
be only able to verbalise the word yes, but was able to description of the music type used. Since music types can
correctly sing familiar hymns, producing both the greatly vary (eg, stimulating vs soothing), their expected
melody and the text of the songs.94 effects on physiological parameters, arousal, and affect

www.thelancet.com/neurology Vol 16 August 2017 657

 Review

regulation differ. Furthermore, most of the reviewed rehabilitation technology using virtual reality or adaptive
studies did not use patient-selected or favourite music. music stimulation systems tailored for motor
Because of the strong emotional components of musical rehabilitation, will play an increasing role in bringing
experience, using patient-selected music could be music to patients with neurological disorders, in hospital,
beneficial since it can be considered more meaningful community, and home environments.
and rewarding to the patient than generic music. Contributors
Additional high-quality intervention studies, particularly AJS, VL, and SS searched and reviewed the literature. AJS created the
large-scale trials such as cluster-randomised, multicentre figures and the table, and AJS and SS wrote the primary manuscript,
which was circulated among the other authors (TS, EA, MT, and VL).
randomised controlled trials, in which the established All authors made substantial additions on the basis of their special areas
music inter­ ventions are embedded into the clinical of interest, which were incorporated into the final manuscript.
rehabilitation practice, would be needed to establish the Declaration of interests
efficacy and feasibility in real-life settings of these We declare no competing interests.
approaches. For better comparability of the studies, it Acknowledgments
would also be important to use common outcome TS received support from the Academy of Finland programme (1277693).
measures, clearly document the type of the intervention SS received support from the Research Grants of Turku University
(active vs receptive) and music used (patient-selected vs Hospital (EVO 8140/2012, 2014). AJS received support from the Finnish
Brain Research and Rehabilitation Foundation, Signe and Ane
experimenter-selected), as well as define the optimal time Gyllenberg Foundation, and Maire Taponen Foundation. VL received
of onset and legnth of the music interventions, support from the National Doctoral Programme of Psychology, Finland.
and determine the long-term duration, if any, of their References
rehabilitation effects. In addition, multimodal studies 1 WHO. World report on ageing and health. 2015. http://apps.who.
combining behavioural outcome measures with neuro­ int/iris/bitstream/10665/186463/1/9789240694811_eng.pdf?ua=1
(accessed April 11, 2017).
imaging and neuroendocrinological markers are needed 2 PricewaterhouseCoopers Health Industries. The annual cost of brain
to determine specific neurophysiological mechanisms disease 2012. 2012. http://pwchealth.com/cgi-local/hregister.cgi/reg/
and effects of various music-based interventions in annual-cost-of-brain-disease-2012.pdf (accessed April 11, 2017).
3 Olesen J, Gustavsson A, Svensson M, Wittchen HU, Jönsson B.
patients with a neurological disorder. The economic cost of brain disorders in Europe. Eur J Neurol 2012;
Analysis of the amount of core therapeutic activities 19: 155–62.
received by patients in one rehabilitation centre in 4 Nudo RJ. Recovery after brain injury: mechanisms and principles.
Front Hum Neurosci 2013; 7: 887.
Canada, such as physiotherapy and occupational therapy,
5 Tomassini V, Matthews PM, Thompson AJ, et al.
suggests that patients who have had a stroke receive only Neuroplasticity and functional recovery in multiple sclerosis.
approximately 60% of the recommended rehabilitation.96 Nat Rev Neurol 2012; 8: 635–46.
It is unclear whether this insufficiency results from 6 Hill NL, Kolanowski AM, Gill DJ. Plasticity in early Alzheimer’s
disease: an opportunity for intervention. Top Geriatr Rehabil 2011;
scarceness of rehabilitation resources or other factors, 27: 257–67.
but is likely to be generalisable to most neurological 7 Reetz K, Tadic V, Kasten M, et al. Structural imaging in the
wards. Thus, there is a need for music interventions that presymptomatic stage of genetically determined parkinsonism.
Neurobiol Dis 2010; 39: 402–08.
are widely available and can easily be implemented with
8 Cramer SC, Sur M, Dobkin BH, et al. Harnessing neuroplasticity
minimal investment. These interventions include for clinical applications. Brain 2011; 134: 1591–609.
self-implemented or caregiver-implemented musical 9 Zeiler SR, Krakauer JW. The interaction between training and
activities, such as music listening, and group-based plasticity in the poststroke brain. Curr Opin Neurol 2013; 26: 609–16.
10 Herholz SC, Herholz RS, Herholz K. Non-pharmacological
musical interventions, such as group singing or dancing. interventions and neuroplasticity in early stage Alzheimer’s disease.
In the future, mobile music applications (eg, music Expert Rev Neurother 2013; 13: 1235–45.
streaming, games), as well as novel music-based 11 Agosta F, Gatti R, Sarasso E, et al. Brain plasticity in Parkinson’s
disease with freezing of gait induced by action observation training.
J Neurol 2017; 264: 88–101.
12 Enzinger C, Pinter D, Rocca MA, et al. Longitudinal fMRI studies:
Search strategy and selection criteria exploring brain plasticity and repair in MS. Mult Scler 2016; 22: 269–78.
13 Baroncelli L, Braschi C, Spolidoro M, Begenisic T, Sale A, Maffei L.
We searched PubMed for papers published in English before Nurturing brain plasticity: impact of environmental enrichment.
April 11, 2017, using the Medical Subject Headings “stroke”, Cell Death Differ 2010; 17: 1092–103.
“brain injuries”, “dementia”, “parkinsonian disorders”, 14 Zatorre RJ, Chen JL, Penhune VB. When the brain plays music:
auditory-motor interactions in music perception and production.
“epilepsy”, and “multiple sclerosis”, combined with “music” or Nat Rev Neurosci 2007; 8: 547–58.
“music therapy” and the keywords “melodic intonation 15 Koelsch S. Brain correlates of music-evoked emotions.
therapy”, “rhythmic auditory stimulation”, “rhythmic Nat Rev Neurosci 2014; 15: 170–80.
auditory cueing”, and “music supported therapy”. Additional 16 Särkämö T, Tervaniemi M, Huotilainen M. Music perception and
cognition: development, neural basis, and rehabilitative use of
references were gathered from reference lists and relevant music. Wiley Interdiscip Rev Cogn Sci 2013; 4: 441–51.
articles. We included only the randomised controlled trials 17 Alluri V, Toiviainen P, Jääskeläinen IP, Glerean E, Sams M, Brattico E.
that applied at least a 1-week intervention, and we focused Large-scale brain networks emerge from dynamic processing of
musical timbre, key and rhythm. Neuroimage 2012; 59: 3677–89.
on scientific literature of the past 10 years, but we also 18 Wan CY, Schlaug G. Music making as a tool for promoting brain
included two older landmark studies. plasticity across the life span. Neuroscientist 2010; 16: 566–77.

658 www.thelancet.com/neurology Vol 16 August 2017

 Review

19 Schlaug G. Musicians and music making as a model for the study 41 Särkämö T, Laitinen S, Numminen A, Kurki M, Johnson JK,
of brain plasticity. Prog Brain Res 2015; 217: 37–55. Rantanen P. Pattern of emotional benefits induced by regular
20 Vaquero L, Hartmann K, Ripollés P, et al. Structural neuroplasticity singing and music listening in dementia. J Am Geriatr Soc 2016;
in expert pianists depends on the age of musical training onset. 64: 439–40.
Neuroimage 2016; 1: 106–19. 42 Särkämö T, Laitinen S, Numminen A, Kurki M, Johnson JK,
21 Hole J, Hirsch M, Ball E, Meads C. Music as an aid for postoperative Rantanen P. Clinical and demographic factors associated with the
recovery in adults: a systematic review and meta-analysis. Lancet cognitive and emotional efficacy of regular musical activities in
2015; 386: 1659–71. dementia. J Alzheimers Dis 2015; 49: 767–81.
22 Magee WL, Clark I, Tamplin J, Bradt J. Music interventions for 43 Särkämö T, Tervaniemi M, Laitinen S, et al. Cognitive, emotional,
acquired brain injury. Cochrane Database Syst Rev 2017; 1: CD006787. and social benefits of regular musical activities in early
dementia: randomized controlled study. Gerontologist 2014;
23 Benjamin EJ, Blaha MJ, Chiuve SE, et al. Heart disease and stroke
54: 634–50.
statistics—2017 update: a report from the American Heart
Association. Circulation 2017; 135: e146–603. 44 Raglio A, Bellandi D, Baiardi P, et al. Effect of active music therapy
and individualized listening to music on dementia: a multicenter
24 van Vugt FT, Kafczyk T, Kuhn W, Rollnik JD, Tillmann B,
randomized controlled trial. J Am Geriatr Soc 2015; 63: 1534–39.
Altenmüller E. The role of auditory feedback in music-supported
stroke rehabilitation: a single-blinded randomised controlled 45 Hsu MH, Flowerdew R, Parker M, Fachner J, Odell-Miller H.
intervention. Restor Neurol Neurosci 2016; 34: 297–311. Individual music therapy for managing neuropsychiatric symptoms
for people with dementia and their carers: a cluster randomised
25 Scholz DS, Rohde S, Nikmaram N, et al. Sonification of arm
controlled feasibility study. BMC Geriatr 2015; 15: 84.
movements in stroke rehabilitation—a novel approach in
neurologic music therapy. Front Neurol 2016; 7: 106. 46 Chu H, Yang CY, Lin Y, et al. The impact of group music therapy
on depression and cognition in elderly persons with dementia:
26 Raglio A, Oasi O, Gianotti M, Rossi A, Goulene K,
a randomized controlled study. Biol Res Nurs 2014; 16: 209–17.
Stramba-Badiale M. Improvement of spontaneous language in
stroke patients with chronic aphasia treated with music therapy: 47 Vink AC, Zuidersma M, Boersma F, de Jonge P, Zuidema SU,
a randomized controlled trial. Int J Neurosci 2016; 126: 235–42. Slaets JP. Effect of music therapy versus recreational activities on
neuropsychiatric symptoms in elderly adults with dementia:
27 Tong Y, Forreider B, Sun X, et al. Music-supported therapy (MST)
an exploratory randomized controlled trial. J Am Geriatr Soc 2014;
in improving post-stroke patients’ upper-limb motor function:
62: 392–93.
a randomised controlled pilot study. Neurol Res 2015; 37: 434–40.
48 Vink AC, Zuidersma M, Boersma F, de Jonge P, Zuidema SU,
28 Särkämö T, Ripollés P, Vepsäläinen H, et al. Structural changes
Slaets JP. The effect of music therapy compared with general
induced by daily music listening in the recovering brain after
recreational activities in reducing agitation in people with dementia:
middle cerebral artery stroke: a voxel-based morphometry study.
a randomised controlled trial. Int J Geriatr Psychiatry 2013; 28: 1031–38.
Front Hum Neurosci 2014; 8: 245.
49 Raglio A, Bellelli G, Traficante D, et al. Efficacy of music therapy in
29 van der Meulen I, van de Sandt-Koenderman WM,
the treatment of behavioral and psychiatric symptoms of dementia.
Heijenbrok-Kal MH, Visch-Brink EG, Ribbers GM. The efficacy and
Alzheimer Dis Assoc Disord 2008; 22: 158–62.
timing of melodic intonation therapy in subacute aphasia.
Neurorehabil Neural Repair 2014; 28: 536–44. 50 Narme P, Clement S, Ehrle N, et al. Efficacy of musical
interventions in dementia: evidence from a randomized controlled
30 Cha Y, Kim Y, Hwang S, Chung Y. Intensive gait training with
trial. J Alzheimers Dis 2014; 38: 359–69.
rhythmic auditory stimulation in individuals with chronic
hemiparetic stroke: a pilot randomized controlled study. 51 Ceccato E, Vigato G, Bonetto C, et al. STAM protocol in dementia:
NeuroRehabilitation 2014; 35: 681–88. a multicenter, single-blind, randomized, and controlled trial.
Am J Alzheimers Dis Other Demen 2012; 27: 301–10.
31 Whitall J, Waller SM, Sorkin JD, et al. Bilateral and unilateral arm
training improve motor function through differing neuroplastic 52 Raglio A, Bellelli G, Traficante D, et al. Efficacy of music therapy
mechanisms: a single-blinded randomized controlled trial. treatment based on cycles of sessions: a randomised controlled trial.
Neurorehabil Neural Repair 2011; 25: 118–29. Aging Ment Health 2010; 14: 900–04.
32 Särkämö T, Pihko E, Laitinen S, et al. Music and speech listening 53 Sung HC, Lee WL, Li TL, Watson R. A group music intervention
enhance the recovery of early sensory processing after stroke. using percussion instruments with familiar music to reduce anxiety
J Cogn Neurosci 2010; 22: 2716–27. and agitation of institutionalized older adults with dementia.
Int J Geriatr Psychiatry 2012; 27: 621–27.
33 Särkämö T, Tervaniemi M, Laitinen S, et al. Music listening
enhances cognitive recovery and mood after middle cerebral artery 54 Lin Y, Chu H, Yang CY, et al. Effectiveness of group music
stroke. Brain 2008; 131: 866–76. intervention against agitated behavior in elderly persons with
dementia. Int J Geriatr Psychiatry 2011; 26: 670–78.
34 Altenmüller E, Marco-Pallares J, Münte TF, Schneider S.
Neural reorganization underlies improvement in stroke-induced 55 Cooke ML, Moyle W, Shum DH, Harrison SD, Murfield JE.
motor dysfunction by music-supported therapy. Ann NY Acad Sci A randomized controlled trial exploring the effect of music on
2009; 1169: 395–405. agitated behaviours and anxiety in older people with dementia.
Aging Ment Health 2010; 14: 905–16.
35 Schneider S, Schoenle PW, Altenmüller E, Muente TF. Using musical
instruments to improve motor skill recovery following a stroke. 56 Guetin S, Portet F, Picot MC, et al. Effect of music therapy on
J Neurol 2007; 254: 1339–46. anxiety and depression in patients with Alzheimer’s type dementia:
randomised, controlled study. Dement Geriatr Cogn Disord 2009;
36 Jeong S, Kim MT. Effects of a theory-driven music and movement
28: 36–46.
program for stroke survivors in a community setting. Appl Nurs Res
2007; 20: 125–31. 57 Pohl P, Dizdar N, Hallert E. The Ronnie Gardiner rhythm and
music method—a feasibility study in Parkinson’s disease.
37 Thaut MH, Leins AK, Rice RR, et al. Rhythmic auditory stimulation
Disabil Rehabil 2013; 35: 2197–204.
improves gait more than NDT/Bobath training in near-ambulatory
patients early poststroke: a single-blind, randomized trial. 58 de Bruin N, Doan JB, Turnbull G, et al. Walking with music is a
Neurorehabil Neural Repair 2007; 21: 455–9. safe and viable tool for gait training in Parkinson’s disease: the
effect of a 13-week feasibility study on single and dual task walking.
38 Schauer M, Mauritz K. Musical motor feedback (MMF) in walking
Parkinsons Dis 2010; 2010: 483530.
hemiparetic stroke patients: randomized trials of gait improvement.
Clin Rehabil 2003; 17: 713–22. 59 Hackney ME, Earhart GM. Effects of dance on movement control in
Parkinson’s disease: a comparison of Argentine tango and
39 Thaut MH, McIntosh GC, Rice RR. Rhythmic facilitation of gait
American ballroom. J Rehabil Med 2009; 41: 475–81.
training in hemiparetic stroke rehabilitation. J Neurol Sci 1997;
151: 207–12. 60 Hackney ME, Earhart GM. Health-related quality of life and
alternative forms of exercise in Parkinson disease.
40 Sánchez A, Maseda A, Marante-Moar MP, de Labra C,
Parkinsonism Relat Disord 2009; 15: 644–48.
Lorenzo-López L, Millán-Calenti JC. Comparing the effects of
multisensory stimulation and individualized music sessions on 61 Hackney ME, Kantorovich S, Levin R, Earhart GM. Effects of tango
elderly people with severe dementia: a randomized controlled trial. on functional mobility in Parkinson’s disease: a preliminary study.
J Alzheimers Dis 2016; 52: 303–15. J Neurol Phys Ther 2007; 31: 173–79.

www.thelancet.com/neurology Vol 16 August 2017 659

 Review

62 Gatti R, Tettamanti A, Lambiase S, Rossi P, Comola M. 81 Schlaug G, Marchina S, Norton A. Evidence for plasticity in
Improving hand functional use in subjects with multiple sclerosis white-matter tracts of patients with chronic Broca’s aphasia
using a musical keyboard: a randomized controlled trial. undergoing intense intonation-based speech therapy.
Physiother Res Int 2015; 20: 100–07. Ann NY Acad Sci 2009; 1169: 385–94.
63 Conklyn D, Stough D, Novak E, Paczak S, Chemali K, Bethoux F. 82 Salimpoor VN, Benovoy M, Larcher K, Dagher A, Zatorre RJ.
A home-based walking program using rhythmic auditory Anatomically distinct dopamine release during anticipation and
stimulation improves gait performance in patients with multiple experience of peak emotion to music. Nat Neurosci 2011; 14: 257–62.
sclerosis: a pilot study. Neurorehabil Neural Repair 2010; 24: 835–42. 83 Okada K, Kurita A, Takase B, et al. Effects of music therapy on
64 Bodner M, Turner RP, Schwacke J, Bowers C, Norment C. autonomic nervous system activity, incidence of heart failure
Reduction of seizure occurrence from exposure to auditory events, and plasma cytokine and catecholamine levels in elderly
stimulation in individuals with neurological handicaps: patients with cerebrovascular disease and dementia. Int Heart J
a randomized controlled trial. PLoS One 2012; 7: e45303. 2009; 50: 95–110.
65 Morris SB. Estimating effect sizes from pretest-posttest-control 84 Bradt J, Dileo C, Potvin N. Music for stress and anxiety reduction in
group designs. Organ Res Methods 2008; 11: 364–86. coronary heart disease patients. Cochrane Database Syst Rev 2013;
66 Stroke Association. State of the Nation. Stroke statistics— 12: CD006577.
January 2016. 2015. https://www.stroke.org.uk/sites/default/files/ 85 Radley J, Morilak D, Viau V, Campeau S. Chronic stress and brain
stroke_statistics_2015.pdf (accessed April 11, 2017). plasticity: mechanisms underlying adaptive and maladaptive
67 Albert ML, Sparks RW, Helm NA. Melodic intonation therapy for changes and implications for stress-related CNS disorders.
aphasia. Arch Neurol 1973; 29: 130–31. Neurosci Biobehav Rev 2015; 58: 79–91.
68 Nys GM, van Zandvoort MJ, de Kort PL, Jansen BP, de Haan EH, 86 Barugh AJ, Gray P, Shenkin SD, Maclullich AM, Mead GE.
Kappelle LJ. Cognitive disorders in acute stroke: prevalence and Cortisol levels and the severity and outcomes of acute stroke:
clinical determinants. Cerebrovasc Dis 2007; 23: 408–16. a systematic review. J Neurol 2014; 261: 533–45.
69 Hackett ML, Pickles K. Part I: frequency of depression after stroke: 87 Nilsson U. Soothing music can increase oxytocin levels during bed
an updated systematic review and meta-analysis of observational rest after open-heart surgery: a randomised control trial. J Clin Nurs
studies. Int J Stroke 2014; 9: 1017–25. 2009; 18: 2153–61.
70 Hall CB, Lipton RB, Sliwinski M, Katz MJ, Derby CA, Verghese J. 88 Nilsson U. The effect of music intervention in stress response to
Cognitive activities delay onset of memory decline in persons who cardiac surgery in a randomized clinical trial. Heart Lung 2009;
develop dementia. Neurology 2009; 73: 356–61. 38: 201–07.
71 Hanagasi HA, Tufekcioglu Z, Emre M. Dementia in Parkinson’s 89 Raglio A, Attardo L, Gontero G, Rollino S, Groppo E, Granieri E.
disease. J Neurol Sci 2017; 374: 26–31. Effects of music and music therapy on mood in neurological
72 Nombela C, Hughes LE, Owen AM, Grahn JA. Into the groove: patients. World J Psychiatry 2015; 5: 68–78.
can rhythm influence Parkinson’s disease? Neurosci Biobehav Rev 90 Pan A, Sun Q, Okereke OI, Rexrode KM, Hu FB. Depression and
2013; 37: 2564–70. risk of stroke morbidity and mortality: a meta-analysis and
73 Rao AK, Mazzoni P, Wasserman P, Marder K. Longitudinal change systematic review. JAMA 2011; 306: 1241–49.
in gait and motor function in pre-manifest Huntington’s Disease. 91 Towfighi A, Ovbiagele B, El Husseini N, et al. Poststroke depression:
PLoS Curr 2011; 3: RRN1268. a scientific statement for healthcare professionals from the
74 Fisher RS. Therapeutic devices for epilepsy. Ann Neurol 2012; American Heart Association/American Stroke Association. Stroke
71: 157–68. 2017; 48: e30–43.
75 Dastgheib SS, Layegh P, Sadeghi R, Foroughipur M, Shoeibi A, 92 Zarate JM. The neural control of singing. Front Hum Neurosci 2013;
Gorji A. The effects of Mozart’s music on interictal activity in 7: 237.
epileptic patients: systematic review and meta-analysis of the 93 Wan CY, Zheng X, Marchina S, Norton A, Schlaug G.
literature. Curr Neurol Neurosci Rep 2014; 14: 420. Intensive therapy induces contralateral white matter changes in
76 Meyer GF, Spray A, Fairlie JE, Uomini NT. Inferring common chronic stroke patients with Broca’s aphasia. Brain Lang 2014; 136: 1–7.
cognitive mechanisms from brain blood-flow lateralization data: 94 Dahlin O. Beräattelse om en dumbe, som kan siumga (On a mute
a new methodology for fTCD analysis. Front Psychol 2014; 5: 552. who can sing). Kungl Svenska Vetensk Acad Handl 1745; 6: 114–15
77 Murphy TH, Corbett D. Plasticity during stroke recovery: from [in Swedish].
synapse to behaviour. Nat Rev Neurosci 2009; 10: 861–72. 95 Jacobsen JH, Stelzer J, Fritz TH, Chetelat G, La Joie R, Turner R.
78 Grau-Sanchez J, Amengual JL, Rojo N, et al. Plasticity in the Why musical memory can be preserved in advanced Alzheimer’s
sensorimotor cortex induced by music-supported therapy in stroke disease. Brain 2015; 138: 2438–50.
patients: a TMS study. Front Hum Neurosci 2013; 7: 494. 96 Foley N, McClure JA, Meyer M, Salter K, Bureau Y, Teasell R.
79 Ripollés P, Rojo N, Grau-Sanchez J, et al. Music supported therapy Inpatient rehabilitation following stroke: amount of therapy
promotes motor plasticity in individuals with chronic stroke. received and associations with functional recovery. Disabil Rehabil
Brain Imaging Behav 2016; 10: 1289–307. 2012; 34: 2132–38.
80 Amengual JL, Rojo N, Veciana de Las Heras M, et al.
Sensorimotor plasticity after music-supported therapy in chronic
stroke patients revealed by transcranial magnetic stimulation.
PLoS One 2013; 8: e61883.

660 www.thelancet.com/neurology Vol 16 August 2017