Is there potential for learning in amusia? A study of the effect of singing intervention in...

Ann. N.Y. Acad. Sci. ISSN 0077-8923

ANNALS OF THE NEW YORK ACADEMY OF SCIENCESIssue: The Neurosciences and Music IV: Learning and Memory

Is there potential for learning in amusia? A study of theeffect of singing intervention in congenital amusia

Susan Anderson,1 Evangelos Himonides,2 Karen Wise,3 Graham Welch,2 and Lauren Stewart11Department of Psychology, Goldsmiths, University of London, London, United Kingdom. 2Institute of Education, University ofLondon, London, United Kingdom. 3Faculty of Music, University of Cambridge, Cambridge, United Kingdom

Address for correspondence: Lauren Stewart, Goldsmiths, University of London-Psychology, Whitehead Building, New Cross,SE14 6NW, United Kingdom. [email protected]

Congenital amusia is a neurodevelopmental disorder of musical perception and production. Much research hasfocused on characterizing the deficits within this special population; however, it is also important from both apsychological and educational perspective to determine which aspects of the disorder may be subject to changebecause this will also constrain theorizing about the nature of the disorder, as well as facilitating possible futureremediation programs. In this small-scale study, a professional singing teacher used a broad-brush interventionapproach with five individuals diagnosed with congenital amusia. The compensatory elements were designed toenhance vocal efficiency and health, singing technique, musical understanding, pitch perception, and production.Improvements were observed in most individuals in perception, indexed via the Montreal Battery for the Evaluation ofAmusia scale subtest and in the vocal performance of familiar songs. The workshop setting gave a unique opportunityfor observation and discussion to inform further investigations of this disorder.

Keywords: congenital amusia; poor-pitch singing; intervention

Introduction

Congenital amusia is a developmental disorderthat impacts negatively on the perception of mu-sic, which has been estimated to occur in about4% of otherwise normal individuals.1 In everydaylife, these individuals have difficulties in detectinganomalous pitches in melodies, judging dissonancein musical excerpts, and recognizing melodies with-out lyrics.2 These deficits are diagnosed and definedin perceptual terms using the Montreal Battery ofEvaluation of Amusia (MBEA).1 Additional featuresreported in these individuals are higher thresholdsthan controls for the discrimination of pitch direc-tion,3 and pitch memory deficits.3,4

The pitch production of those diagnosed withcongenital amusia is less commonly assessed, butthere is a general picture of impairment: nearly allamusics are unable to sing in tune.2 However, cleardifferentiation is made in the literature between con-genital amusia and “tone-deafness,” a self-reportedproblem that almost always carries the meaning of“can’t sing.”5 Poor-pitch singing is relatively com-mon; about 15% of typical populations consider

themselves “tone-deaf,” whereas musical perceptualdifficulties as diagnosed by the MBEA occur less fre-quently.6 The phenomenon of poor-pitch singingwithout accompanying perceptual deficits has beenreferred to as “purely vocal tone-deafness.”7 Poor-pitch singing has been studied in educational termssince the 19th century from both perceptual andproduction perspectives.8 Production deficits maybe attributable to a variety of causes, both physicaland cognitive, in addition to perceptual causes, suchas poor vocal-motor planning and adjustment, lackof tonal schema, reduced vocal flexibility, misun-derstanding of speech/singing differentiation,9 andis most likely to arise from a stagnation of vocal de-velopment arising from childhood self-perceptionas a “nonsinger.”10

Counterintuitively, proficient singing despite im-paired perception has been noted in several stud-ies of amusia.2 A mismatch between implicit andexplicit pitch awareness has been observed in be-havioral and neuroimaging studies,12 leading to thepossibility that there may be residual implicit pitchabilities in the amusic population. These factors,coupled with the confirmation that neural plasticity

doi: 10.1111/j.1749-6632.2011.06404.xAnn. N.Y. Acad. Sci. 1252 (2012) 345–353 c© 2012 New York Academy of Sciences. 345

Is there potential for learning in amusia? Anderson et al.

is now known to be lifelong,13 and that any poten-tial improvements in production may also benefitperception according to premotor theories of per-ception,14 motivated a singing intervention study inwhich both perception and production were moni-tored. From a pedagogical perspective, many singingteachers have anecdotal evidence of major improve-ments in singing ability with very challenged singers,and are often of the opinion that anyone can betaught to sing.

In this study, a wide range of activities were usedwith five individuals with congenital amusia, in anattempt to improve both pitch production and po-tentially also perception. The program was devisedand delivered in seven weekly singing workshops bya professional singing teacher with many years ofexperience working with poor-pitch singers. It fo-cused on vocal flexibility, musical awareness, devel-opment of tonal schema, and used “Sing and See,”15

a real-time visual feedback program,16 that allowedparticipants to visualize the pitches they produced.The five amusic participants were matched with fivenonamusic controls on gender, age, years of educa-tion, and years of musical training; all participantswere asked to complete a “musical experience” chart,highlighting positive and negative musical experi-ences during their lives. Although the controls wereassessed only once, the amusics were assessed pre-and postintervention.

Hypothesis

It was hypothesized that an intervention program,in the form of singing workshops, would improvethe pitch production abilities and potentially alsoperception abilities of the amusic participants.

Method

DesignThe design was a classic intervention study, withassessments made before and after a group of fiveamusic participants had taken part in seven weeklysinging workshops. It was both a within-subjects de-sign, comparing assessment results pre- and postin-tervention, and a between-subjects design, as theamusic group results, preintervention, were alsocompared with those of matched controls.1

ParticipantsFive amusic female adults (mean age 54.8) and fivefemale matched controls (mean age 58.0) partici-

pated in the study. Participants were drawn from apool of amusics and controls recruited via an on-line version of the scale subtest from the MBEA,and subsequently tested on four subtests of the bat-tery under controlled laboratory conditions. Partic-ipants had also undergone audiometry tests, as wellas completing National Adult Reading Test and digitspan tests. Status as amusic or nonamusic was basedon a composite score across the three pitch-basedsubtests of MBEA.1 A summary of relevant partic-ipant characteristics is given in Table 1. Informedconsent was obtained from all participants.

Materials and procedure

Assessment sessionsAll testing took place in a sound-attenuated booth.The amusic participants were tested individuallyacross two assessment sessions, once before the in-tervention workshops began, and once after the lastworkshop session. The control group was testedin a single assessment session. All assessment ses-sions followed a protocol devised previously by anexperienced singing teacher.17 Each session beganwith basic vocal tasks such as exaggerated speechpatterns, glides, and stepwise exploration of vocalrange, which served as a warm-up and familiar-ized participants with the researcher and the testingsituation.

The actual assessment sessions included both per-ception and production tasks. In terms of percep-tion, performance was measured on the MBEA scalesubtest (online) as well as on a computer-basedpitch-matching task (details below). In terms ofproduction, singing accuracy was measured usinga standard, well-known song (“Happy Birthday”)as well as a self-chosen song, and the ability to imi-tate and match sequences of tones, which ranged inlength from one to five notes.

Computer pitch-matching taskThe computer pitch-matching task (CPM) was de-signed to allow participants to engage in active pitchmatching, but without the requirement to use thevoice. They were instructed to use two keys on acomputer keyboard to play first a “target” tone andthen a “comparison” tone, presented through head-phones. By turning a dial clockwise to raise pitchand anticlockwise to lower it, the pitch of the com-parison note could be brought into alignment withthat of the target tone. Participants were able to

346 Ann. N.Y. Acad. Sci. 1252 (2012) 345–353 c© 2012 New York Academy of Sciences.

Anderson et al. Is there potential for learning in amusia?

Table 1. Matching of five amusics with five controls on all preintervention measures

MBEA Song VPM

Musical pitch CPM rating average

Age Education training composite (cents average (cents

Group N (years) (years) (years) (out of 90) deviation) (1–8) deviation)

Amusic 5 � 53.80 16.60 2.80 55.20 25.66 3.04 256

� 9.70 0.89 4.70 6.98 9.45 1.50 136

Control 5 � 57.00 15.80 0.80 81.20 33.83 6.37 43

� 3.80 1.64 1.30 6.98 27.40 0.58 7

U 10.50 10.00 9.50 <0.001 10.00 0.50 <0.001

Z –0.42 –0.57 –0.67 –2.61 –0.52 –2.53 –2.61

P 0.69 0.69 0.55 0.01 0.69 0.01 0.01

r –0.39 –0.18 –0.19 –0.83 –0.17 –0.80 –0.83

play only one tone at any one time, but they couldswitch between target and comparison notes at will.Target notes spanned the chromatic scale from G3

to G4, and comparison tones were a semitone (100cents), major 3rd (400 cents), or a perfect 5th (700cents) above or below the target note. The pitchadjustments made by turning the dial were sampledapproximately every 93 milliseconds. The difference(in cents) between the final pitch of the comparisontone and the target tone was calculated as a measureof perceptual pitch-matching accuracy.

Singing tasksSongs. Participants were asked in advance of thefirst assessment to think of a song that they knewwell, which they would be willing to sing to the re-searcher. No starting note was given for their ownchoice song, allowing each participant to choosea comfortable singing range. This was followed bytwo performances of “Happy Birthday,” first with-out, then with, accompaniment. The starting pitchchosen for the accompanied version was dependenton the vocal range established in the warm-up, andthe accompaniment added simple harmony as wellas doubling the vocal line. Sung performances wererecorded using Audacity, extracted from the Au-dacity files, and copied in random order onto CD.Evaluation was carried out by two independent ex-perts, using an accuracy rating scale developed foruse with adults by Wise and Sloboda.17

Vocal pitch matching (echo and synchronizeconditions). The stimuli used for this task weretaken from a previous study,17 using single pitches

and meaningful melodic fragments. The stimuli,recorded by an accurate but untrained female singer,were heard in two separate blocks, each consist-ing of single notes, pairs of notes, groups of threenotes, and finishing with groups of five-note pat-terns. In the first block (echo condition), the par-ticipant heard the stimulus and was asked to sing itback in their own time; in the second block (syn-chronize condition), a click track cued the partici-pant’s synchronized response. All notes and patternswere in D major, in the range B3 to B4, using sim-ple tuneful relationships and avoiding any singleinterval larger than a perfect 5th. They were ana-lyzed for pitch accuracy using Praat, with the fre-quency of the steadiest portion of each note beingrecorded.

Intervention phaseThe five participants met with the researcher (SA)for seven weekly group-singing workshops, eachlasting one and a half hours, with worksheets andCDs providing guided practice between sessions.The structure and content of the workshops wasdesigned to remediate deficits identified from theliterature and from pedagogical experience as pos-sibly underlying the phenomenon of amusia. Theywere intended to be relaxed, fun, nonjudgmental,vocally and intellectually stimulating, and aimed toimprove vocal confidence and efficient voicing—regardless of the research outcome.

All workshops began with vocal warm-ups thatwere not pitch specific, such as lip-flutters (motor-bike sounds), glides on /u/, and a “fun” physical andvocal warm-up that incorporates the basic children’s

Ann. N.Y. Acad. Sci. 1252 (2012) 345–353 c© 2012 New York Academy of Sciences. 347


chant featuring the falling minor 3rd (scale pitches53 653). Elements that contribute to strong, efficientvoicing were reinforced, such as use of the trans-verse abdominis muscles to support the vocal sound,derived from Accent Method techniques,18 and anopen, unconstricted laryngeal posture.19 Simple vo-cal anatomy was introduced, with anatomical andphysiological explanations given whenever appro-priate. An additional intervention to simulate pitchperception was the playing “by ear” of simple well-known melodies on pianos/keyboards in the depart-ment, and free composition.

Between sessions, participants were asked to prac-tice three or four times a week, for about 15 minper session, using specially prepared CDs and pro-gressive exercises with Sing and See. The final feed-back forms indicated compliance with these re-quirements. The exercises progressed from glidesthroughout the vocal range, to steady sounds onspecified “target pitches” around middle C (C4),a comfortable note for all participants. The de-scending minor 3rd (5–3) was introduced in thechildren’s chant mentioned above, before descend-ing major triads, 5 3 1. The use of numbers wasused consistently to encourage awareness of pitchrelationships.

Results

Owing to the small number of participants involvedin this study (five participants in each group), non-parametric statistics were appropriate; the Mann-Whitney test for two unrelated samples was used tocompare amusic participants’ preintervention withcontrols, and the Wilcoxon signed-rank test wasused for pre- and postintervention comparison. Inboth cases, effect sizes can be more meaningfully in-terpreted than P values.20 Following group compar-isons, first amusic preintervention (amusic1) versuscontrols, and then amusics pre- versus postinter-vention (amusic1 and amusic2), we focus on someof the individual profiles within the amusic groupdata.

Preintervention amusics and controls

PerceptionBy definition, the MBEA pitch composite score wassignificantly different between the amusic and con-trol groups before intervention (amusic1 [Mdn =56] and controls [Mdn = 83] giving U = 0.000,P = 0.008, r = 0.83 [large effect size]). In contrast,

there was a lack of group differences on the CPMtask (amusics Mdn = 31.4, controls Mdn = 27.1,U = 10, P = 0.65, r = –0.17 [small effect size];Table 1).

ProductionIn all singing tasks, controls scored significantlyhigher than the amusic group before intervention(U = 0.5, z = –2.530, P = 0.008, r = 0.8 [large ef-fect size]), although one amusic (HR) showed a levelof accuracy that was almost comparable to that seenin the control group. In vocal pitch matching, amu-sics were significantly less accurate than the controlgroup in both conditions, displaying a wider spreadof scores and, in the echo condition, there was agreater decrement in performance as the sequencesincreased in length (Fig. 1).

Amusic results pre- and postintervention

PerceptionIn comparing the online scale subtest of the MBEAbefore and after the intervention period, we ob-served no significant improvement at a group level,but important individual changes that contributedto a large effect size (amusic1 Mdn = 17, amusic2Mdn = 21, z = 1.5, P = 0.19, r = –0.51 [large effectsize]). Improvement on the scale subtest was seen inall but one amusic participant. There were no signif-icant changes in the computerized pitch-matchingscores, which, even preintervention, were similar tothe control scores (Table 2).

ProductionThe intervention resulted in the greatest improve-ments for the song ratings, seen in all songs (ownchoice, “Happy Birthday” solo, and “Happy Birth-day” accompanied). In terms of changes in perfor-mance at a group level, performance of one’s ownchoice song yielded the least improvement, whereas“Happy Birthday” (accompanied) showed the mostpositive change (Fig. 2). The vocal pitch-matchingtask showed little significant change, although therewas a large effect size nearing significance for thesynchronized condition when scores were averagedacross all sequence lengths (z = –1.826, P = 0.068,r = –0.61).

Individual profilesWith such a small and heterogeneous group, not-ing individual differences in results is very relevantfor the integrity of the study (Table 2). One of the



Figure 1. Vocal pitch matching (echo condition) in amusics preintervention and controls. The figure shows the effect of sequencelength as deviation in cents from the target pitch.

participants who showed the weakest performancein all baseline conditions (ABL) reported hearingvery little music in her preschool years, as did JMand JP (amusics) and one control. After interven-tion, ABL’s online MBEA score improved from 15to 21, and her song rating for “Happy Birthday” (ac-companied) saw the highest single change, from 2/8to 6/8. JM began from a low MBEA score and thelowest singing rating. After the intervention, mod-est improvements were made in both conditions ofthe vocal pitch-matching task, with minimal posi-tive change in song ratings. Participant JP also madeseveral positive changes from a low baseline score;her MBEA online score improved by four points,her song rating average increasing by 0.66, and vo-cal pitch-matching performance was also more ac-curate, postintervention, in both conditions.

At the preintervention assessment, HR’s produc-tion profile bore more resemblance to the con-trol group than the amusic group. Her scores im-proved in all tasks postintervention, taking heronline MBEA score from 23 to 28, increasing heraverage song ratings from 5.67 to 6.17, and reduc-ing the deviation in VPM from 78 to 42 cents. Hercomputerized pitch-matching score remained con-sistent with an average small deviation of 10 cents(1/10 of a semitone). ABK made a small improve-ment in song ratings, but her performance worsenedfor both MBEA score and vocal pitch matching. In

particular, this participant was unable to completethe vocal pitch-matching task (synchronized condi-tion), remarking that she was distressed by it going“too fast.”

Discussion

This study represents an exploratory investigationinto the potential for change in musical behavior ina small group of individuals with congenital amu-sia. Although genetic findings21,22 indicate that theremay be innate factors contributing to the disorder,recent research regarding the potential for learn-ing and plasticity to occur throughout the lifespan13

suggests that behavioral changes may indeed be pos-sible. Despite the limitations of group size and shortintervention period, some positive changes wereseen in both perception and/or production abili-ties in three of the five individuals, implying thatcongenital amusia may not be entirely immune toenvironmental modification. Below, we consider themain findings of the study and discuss their rele-vance to theories of the disorder.

Profiles of performance in congenitalamusia

The five participants exhibited different pro-files of perception and production at the outsetacross the measures we assessed. For example, JMshowed consistently accurate pitch matching in the



Table 2. Amusics, individual performances on all measures pre- and postintervention

MBEA CPM Song ratings Own Happy Happy VPM average VPM echo VPM sync

online (cents average choice Birthday Birthday (cents (cents (cents

(/30) deviation) (1–8) Solo (1–8) Acc (1–8) (1–8) deviation) deviation) deviation)

Preintervention

ABL 15 27 2.67 3.5 2.5 2.0 312 234 390

ABK 21 34 2.67 3.5 2.0 2.5 213 254 173

HR 23 10 5.67 5.0 6.0 6.0 78 103 54

JM 15 26 2.00 2.0 2.0 2.0 447 332 562

JP 17 33 2.17 2.5 2.5 1.5 231 225 238

� 18.20 25.66 3.04 3.30 3.00 2.80 256.20 229.60 283.40

� 3.63 9.45 1.50 1.15 1.70 1.82 135.81 82.39 137.39

Postintervention

ABL 21 32 4.17 3.5 3.0 6.0 313 265 351

ABK 19 41 3.50 4.0 3.0 3.5 357 357 N/A

HR 28 10 6.17 6.0 5.0 6.5 42 47 38

JM 17 26 2.17 2.0 2.0 2.5 315 274 357

JP 21 59 2.83 2.0 4.0 2.5 127 116 138

� 21.2 28.21 3.77 3.50 3.40 4.20 230.80 211.80 221.00

� 4.15 11.97 1.54 1.66 1.14 1.92 138.02 126.61 158.93

z –1.63 –1.10 –2.02 –0.82 –0.92 –2.04 –0.41 –0.67 –1.83

P 0.10 0.27 0.04 0.41 0.36 0.04 0.69 0.50 0.07

r –0.51 –0.35 –0.64 –0.26 –0.29 –0.65 –0.13 –0.21 –0.61

computerized perception task, but her sung imi-tation of novel pitch patterns and familiar songswas poor, whereas ABK had poor pitch perception,coupled with poor sung performance. In contrast,participant HR performed equally as well as con-trols in all measures. Such variance in profile istypical across developmental disorders in generaland suggests that different individuals may benefitfrom different aspects of a broad-brush interventionapproach. Overall, there was less change in percep-tion measures pre- and postintervention, whereasproduction scores improved for all participants inthe song ratings, and for some in the vocal pitch-matching task as well.

MBEA and CPM tasks

The finding of improved scores on the scale subtestof the MBEA in all but one participant is notewor-thy. Although HR scored 23 (above the cutoff forthat particular subtest preintervention),1 she wasconsidered amusic on the basis of a composite scoreacross all three pitch subtests. The average change in

score was three points, with participant ABL mak-ing a substantial positive change of six points; HR,five points; JP, four points; and ABK making a neg-ative change of two. Although a small increase inscore may be attributed to a test–retest factor, sucheffects are typically modest and would be unable toaccount for the magnitude of improvement seen inthree participants. As the test is specifically found tobe suitable to “measure evolution of performanceafter recovery or therapy,”1 these positive changescan be viewed as individual improvements in musi-cal perception.

The computerized pitch-matching program wasdesigned to test the ability of participants to de-tect, compare, and match small differences in pitch,while minimizing the potential impact of memory.Accurate performance in individuals with amusiais consistent with findings of typical pitch detec-tion thresholds (approximately 0.2 semitones4) inthe context of psychophysical tasks requiring par-ticipants to indicate which of three tones is the “oddone out.”3 This study showed that individuals with



Figure 2. Ratings for the singing of amusic participants, preintervention (amusic t1) and postintervention (amusic t2). Perfor-mances were rated on a scale of 1–8, 2 indicating that words and rhythm were mainly correct, but pitches were random and therewere errors in contour; and 6 indicating that key is maintained throughout and that the melody is fairly accurately represented.

amusia could not only detect small pitch differences,but could actively match two differing tones, rely-ing on an iterative process of feedback and compar-ison. Performance on this task did not change sig-nificantly following the intervention, being alreadyequal to that of controls.

Singing and vocal imitation tasks

The most notable improvements in production ac-curacy occurred in line three of “Happy Birthday,”a phrase consisting of an octave leap followed by adescending major triad; postintervention, this wassung accurately by ABL in the accompanied con-dition and JP in the solo condition. These shapes,but not in this context, were featured in the inter-vention as basic building tools of musical patterns,essential for the development of tonal schema. Theimprovements reported suggest that this element ofmusical understanding had been absent before theintervention. The pedagogical expectation that sup-port (either accompaniment in “Happy Birthday” orsynchronized sounds in vocal pitch-matching tasks)

would improve intonation is true of self-defined“tone-deaf” individuals,17 although Hutchins et al.found that the performance of amusic participantsin a pitch-matching trial was not affected by feed-back condition.23 There were mixed responses re-garding the role of support, a finding echoing thatof Wise,9 with the two participants above (ABL andJP) consistently having more pitch deviation in vocalpitch matching (synchronized condition) comparedwith vocal pitch matching (echo condition). Timingconstraints in the synchronized condition may havehad a negative impact on these individuals in termsof vocal production as they had very little experiencewith singing control and coordination. This coordi-nation typically develops in childhood, with roughly30% of children aged seven unable to sing accurately,reducing to about 4% four years later.7 Developmentis often curtailed when children are criticized dur-ing these formative years, and they then stop tak-ing part in singing activities. Many adults attend-ing groups for nonsingers recall an abiding senseof “deep emotional upset” associated with their



childhood singing experiences, subsequently avoid-ing social situations involving singing.13 In the feed-back document completed after the second assess-ment session, all amusic participants indicated thatSing and See had been one of the most importantelements of the intervention. They also commentedinformally that they now listened to music insteadof ignoring it.

It is notable that three amusic participants, ABL,JM, and JP, reported no preschool musical expe-rience. Although it has been suggested that pitchprocessing hinges upon the need to experience spec-trally and temporally rich sound to wire brain cir-cuits for pitch processing,24 one cannot assume thatthis lack of musical input was responsible for a fail-ure to develop musical awareness in the normal waybecause one of the control groups was also sim-ilar in this regard. However, it is certainly possiblethat musical difficulties arising because of biologicalfactors were exacerbated by the lack of musical stim-ulation in early childhood. Lack of musical stimulusin childhood may happen for many reasons, includ-ing those of parental choice and lack of electricitycited by three of the amusic participants, but seemsless likely to occur now than 40 to 50 years ago,when many of the participants were growing up. In-deed, the decreasing presence of congenital amusiain younger generations may be because of the greatlyincreased “musical stimulation and experience” incontemporary life.25

Conclusions

A workshop environment allowed close observa-tion of individuals and their relationship withmusic and singing over a sustained period oftime. The small numbers and heterogeneity of thegroup made statistical analysis difficult, highlight-ing individual variations rather than illuminat-ing overall “amusic” group characteristics. How-ever, changes of varying degrees were found in thesong production of all amusic participants, andall except one participant improved in their on-line MBEA score. The research project has sug-gested both productive compensatory tactics for thesinging teacher working with poor-pitch singers,and fruitful lines for further laboratory-basedresearch.

Conflicts of interest

The authors declare no conflicts of interest.

References

1. Peretz, I., A.S. Champod & K. Hyde. 2003. Varieties of musi-cal disorders–The Montreal Battery of Evaluation of Amusia.Ann. N.Y. Acad Sci. 999: 58–75.

2. Dalla Bella, S., J.-F. Giguere & I. Peretz. 2009. Singing incongenital amusia. J. Acoust. Soc. Am. 126: 414–424.

3. Foxton, J.M., J.L. Dean, R. Gee, et al. 2004. Characterisationof deficits in pitch perception underlying ‘tone deafness’.Brain 127: 801–810.

4. Willliamson, V.J. & L. Stewart. 2010. Memory for Pitch incongenital amusia: beyond a fine-grained pitch discrimina-tion problem. Memory 18: 657–669.

5. Sloboda, J., K.J. Wise & I. Peretz. 2005. Quantifying tonedeafness in the general population. Ann. N.Y. Acad Sci. 1060:255–261.

6. Cuddy, L.L., L.-L. Balkwill, I. Peretz & R.R. Holden. 2005.Musical difficulties are rare: a study of ‘tone-deafness’among university students. Ann. N.Y. Acad Sci. 1060: 311–324.

7. Dalla Bella, S., J.-F. Giguere & I. Peretz. 2007. Singing pro-ficiency in the general population. J. Acoust. Soc. Am. 121:1182–1189.

8. Welch, G.F. 1979. Poor-Pitch singing: a review of the litera-ture. Psychol. Music 7: 50–58.

9. Wise, K.J. 2009. Understanding ‘Tone-Deafness’: A Multi-Compositional Analysis of Perception, Cognition, Singingand Self-Perception in Adults Reporting Musical Difficulties.Unpublished PhD thesis, Keele University.

10. Welch, G.F. 2006. Singing and vocal development. In TheChild as Musician. G.E. McPherson, Ed.: 311–329. OxfordUniversity Press. Oxford.

11. Loui, P., F.H. Guentehr, C. Mathys & G. Schlaug. 2008.Action-perception mismatch in tone-deafness. Current Bi-ology 18: R331–R332.

12. Peretz, I., E. Brattico, M. Jarvenpaa & M. Tervaniemi. 2009.The amusic brain: in tune, out of key, and unaware. Brain132: 1277–1286.

13. Pellicciari, M.C., C. Miniussi, P.M. Rossini & L. De Gennaro.2009. Increased cortical plasticity in the elderly: changes inthe somato-sensory cortex after paired associative stimula-tion. Neuroscience 163: 266–276.

14. Brown, S. & M.J. Martinez. 2007. Activation of premotorvocal areas during musical discrimination. Brain Cogn. 67:59–69.

15. Wilson, P.H., K. Lee, J. Callaghan & C.W. Thorpe. 2008.Learning to sing in tune: does real-time visual feedback help?J. Interdiscipl. Music Stud. 2: 157–172.

16. Welch, G.F., D.M. Howard & C. Rush. 1989. Real-time vi-sual feedback in the development of vocal pitch accuracy insinging. Psychol. Music 17: 146–157.

17. Wise, K.J. & J.A. Sloboda. 2008. Establishing an empiricalprofile of self-defined ‘tone-deafness’: perception, singingperformance and self-assessment. Musicae Scientae 12: 3–26.

18. Chapman, J.L. 2006. Singing and Teaching Singing: A HolisticApproach to Classical Voice. Plural Publishing Inc. San Diego,Oxford, Brisbane.

19. Shewell, C. 2009. Voice Work: Art and Science in ChangingVoices. Wiley-Blackwell. Chichester.



20. Field, A. 2005. Discovering Statistics Using SPSS. Sage Publi-cations. London, Thousand Oaks, New Delhi.

21. Drayna, D., A. Manichaikul, M. de Lange, et al. 2001. Geneticcorrelates of musical pitch recognition in humans. Science291: 1969–1972.

22. Peretz, I., S. Cummings & M. Dube. 2007. The geneticsof congenital amusia (tone deafness): a family aggregationstudy. Am. J. Hum. Genet. 81: 582–588.

23. Hutchins, S., J.M. Zarate, R.J. Zatorre & I. Peretz. 2010.An acoustical study of vocal pitch matching in congenitalamusia. J. Acoust. Soc. Am. 127: 504–512.

24. Trainor, L.J. 2005. Are there critical periods formusical development? Dev. Psychobiol. 46: 262–278.

25. Peretz, I. 2008. Musical disorders: from behaviour to genes.Curr. Direct. Psychol. Sci. 17: 329–333.


Is there potential for learning in amusia? A study of the effect of singing intervention in...

Documents

Transcript of Is there potential for learning in amusia? A study of the effect of singing intervention in...