Singing ability and disability IntroducJon Professor Graham F Welch UCL InsJtute of EducaJon, London Zatorre & Salimpoor (2013) Loui (2013) Musical development begins pre-birth Russian Dolls model Acous,c links Prosodic and melodic features of mother s voice (speaking & singing) are perceived in utero foetus mother Mother s emojonal state when vocalising (speaking & singing) is encoded hormonally in the filtered interfacing of the mother s and foetus bloodstreams Hormonal links (Welch, 2005; Welch & PreJ in press) Musical development = Basic human design for learning in a variety of socio-cultural contexts Proximal processes for the realisajon of genejc potenjal (cf social ecology/bioecology theory e.g. Bronfenbrenner & Ceci, 1994) (Welch, 2006; 2016) 1
What do we know about musical behaviour? Musical behaviour derives from the interfacing of neuropsychobiological potenjal with experience Musical behaviour develops in a nurturing environment By intenjon and by accident/serendipity tc.columbia.edutc.columbia.edu Early VocalisaJon MulJ-potenJality of early vocalisajon for neuropsychobiological development, especially linguisjc and musical development Parents (and siblings) tend to ujlise musical elements to support the acquisijon of speech in infants (Papousek, 1996:50) Mothers as singing mentors Mothers sing expressively and disjncjvely whilst caring for infants In earliest months, singing is for emojonal regulajon Infants posijve response encourages mothers to sing more Mothers singing is inijally intuijve But intuijve singing contains didacjc elements of pitch and temporal structure Face-to-face context enables combinajon of visual and vocal gestures IniJally, infants make single syllable contribujons to mother-infant duets, progressing to whole phrases and then whole songs (Trehub & Gudmundsdocr, 2014) Toddlerhood (Yen-Ting Wu, 2015) Singing a song from Frozen as seen on YouTube Mother s diary entries over 24 weeks of her 4y+ child s musical behaviours, translated onto the Sounds of Intent in the Early Years framework 2
Incidental benefits of shared music making in the home? InvesJgaJon of parent-child home music making in 3,031 children in Australia (Longitudinal Study of Australian Children: Growing Up in Australia) Parent-child music acjvijes in the early years include: joint and supported singing (including acjon songs, counjng songs, nursery rhymes, and children s songs), generajng original songs to accompany roujne acjvijes, dancing, playing basic instruments, and listening to music on CD, DVD, and MTV Frequency of home music making in 2-3yo had measureable impact subsequently at age 4-5yo Higher frequency of home music acjvijes contributed significantly to children s vocabulary, numeracy, ajenjonal and emojonal regulajon, and prosocial skills (Williams, Barrej, Abad, Broughton & Welch, 2015) Key concepts Basic neuropsychobiological design and maturajonal development are mediated by socio-cultural imperajves (Altenmuller, 2004; McPherson, 2006; Welch, 2006) ParJcular musical behaviours become more or less developed (Welch, 2000; cf Sternberg et al, 2000; Eraut, 2004) Learning happens within a social process (Engeström, 2001; Welch, 2007) General singing ability in the adult populajon? So what counts as normal singing development? And are some people always tone-deaf? The majority of adult individuals can carry a tune (Dalla Bella, Giguère, & Peretz, 2007; Pfordresher & Brown, 2007; Pfordresher, et al., 2010) 10-20% of the general populajon is inaccurate when singing a melody from memory or in pitch-matching tasks (Dalla Bella & Berkowska, 2009; Dalla Bella, Giguère, & Peretz, 2007; Pfordresher & Brown) Occasional singers tended to sing at a faster tempo and with more pitch and Jme errors relajve to professional singers.but.when asked to sing more slowly, they are as accurate as professional singers (Dalla Bella, Giguère, & Peretz, 2007; Larrouy-Maestri & Morsomme, 2014) 3
Neuropsychobiological design and music: the bodymind (Pert, 1986; Thurman & Welch, 2000; Welch, 2005; Welch & PreJ, 2015) Music is an emo,onal experience The brain has an integrated neurological modularity For example these parts of the brain are involved in the analyses of human voices endocrine nervous immune Image by Tsiaras (2005) Belin et al (2000) Voice-selecJve areas in the human auditory cortex. Nature. 403: 309-312. A modular model of music processing in singing AcousJc input (Welch, 2005; Welch & PreJ 2015; adapted from Peretz & Coltheart, 2003) Tonal encoding Pitch organisa,on EmoJon expression analysis Interval analysis Vocal plan formajon Song melody Contour analysis Rhythm analysis AcousJc analysis Temporal organisa,on Musical lexicon Motor analysis AssociaJve memories AcousJc-tophonological conversion Phonological lexicon Song lyrics Tonal encoding Pitch organisa,on Interval analysis 3. Children s singing achieves tonal stability (maintaining key). Contour analysis 2. Children become more accurate in singing the consjtuent intervals within musical phrases. (Welch, in Wise 2015) 1. Children learn to reproduce the contour of melodies correctly (moving the pitches in the correct direcjon). N.B. There may be two separate pitch processing pathways, i.e. For pitch contour (shared by singing and speech) For more precise musical scale-type behaviour (for singing only) (Zatorre & Baum, 2012) 4
A: Overt singing only S1 M1 Vocal students Opera singers singing network Singers vs non-singers Prior to singing lessons A neurological perspecjve: AcJvity changes brain funcjon B: Imagined singing only S1 M1 opera singers vs students C: Imagined more than overt singing singing pracjce Ater one year of singing lessons Parsons & Mithen, 2008 fronto-parietal network (Kleber & Zarate, 2014) An integrated neurological modularity: music and dance Children s speaking Singing (Kleber et al, 2007) ImprovisaJon during jazz (Limb & Braun, 2008) Dance (tango) (Brown et al, 2006) 5
Spoken pitch overview Spoken pitch lowers by approximately 3/4 of a tone from age 5+ to 10+ i.e., the vocal folds elongate and pitch goes down Girls have higher average speaking voices (by 2/3 of a semitone) indicajng slightly smaller vocal folds, on average, compared to boys Asian children have slightly smaller vocal folds, on average, compared to other major ethnic groups (and white < black) but only one semitone difference covers all three ethnic groups (Welch et al 2009) vocal folds Key features of children s singing Early Childhood v 2-year-olds repeat brief phrases with identifiable rhythmic and melodic contour patterns (Dowling, 1999) v 3-year-olds have vocal interplay between spontaneous improvisation and selected elements from the dominant song culture LinguisJc topology (lyrics) Rhythmic surface General tendency in Western cultures for a developmental bias within children s song learning (literature survey) Pitch contour v termed pot-pourri songs (Moog, 1976) v outline songs (Hargreaves, 1996) (Welch, 1994) key stability 6
Three-year longitudinal research data n=184 (boys 87; girls 97) 7 6 Age 5 Age 6 Age 7 5 4 3 2 Leverhulme Project data 1990-1994 1 0 words psycho-acousjc features (pitch, loudness, durajon, Jmbre) single pitchessimple glides fragments songs (pitch) syntacjc and communicajve elements (the grammar of music) structures (detecjng/construcjng pajerns, regularijes) (Welch, Sergeant & White, 1996; 1997; 1998; Welch, 2000b; 2000c; 2004) Ratings for Songs and Test Items by Sex n=184 7 female male 6 Longitudinal data: 5 Children aged 5, then 6, then 7 r 4 a t i n g 3 2 Matching song pitches without words Versus Matching song pitches with words NaJonal Singing Programme Sing Up data 2007-2011 1 = differenjated impact of task by sex 0 5y 6y 7y 5y 6y 7y pitches alone songs (Welch, Sergeant & White, 1996; 1997) 7
18# 16# 14# 12# 10# 8# 6# 4# 2# 0# 12.65# 14.28# 15.24# 15.34# Sing Up is a mulj-faceted programme designed to foster the singing development of Primary school-aged children Schools, Pupils and Assessments (2007-2011 October) 184 schools 11,258 children (52% female) (Primary school aged) 13,096 individual assessments of children s singing and vocal behaviour hjp://www.singup.org/ 40m 2007-2011 4m 2011-2012 April 2012 New launch plus: survey of 1,046 teachers and community musicians micro-analyses of 48 classroom-based singing lessons (Dataset to 1 October 2011) Comfortable mean singing range and age Children s Singing Development: Impact of Sing Up 100# Singing%Development%and%Chronological%Age%(SingUp%7vs7%Non%SingUp)% N=13096% 90# R²#=#0.82189# Normalised#Singing#Development#Score# 80# 70# 60# R²#=#0.60288# Non#SingUp# SingUp# Linear#(Non#SingUp)# Linear#(SingUp)# Age 7 Age 10 g 3 to c 5 (approximately f 3 to e 5 b 11.11# one and a half octaves) (almost two octaves) significant difference for age (p<.001), except between School Years 5 and 6 significant difference between sexes, f > m (p<.001) Welch et al, 2010 mean%comfortable%singing%range%in%semitones% Year#2# Year#3# Year#4# Year#5# Year#6# 50# 40# 4# 5# 6# 7# 8# 9# 10# 11# 12# Pupils'#Chronological#Age#(Years)# Overall, approximately 2 years advantage in singing development for Sing Up parjcipants (Welch et al, in press) 8
Children s actudinal quesjonnaire: 6 themes Children s Actudinal data 1. IdenJty as a singer (emojonal connecjon with singing) 2. IdenJty as a singer (self-efficacy) 3. Singing at home 4. Singing at school 5. Singing in informal secngs 6. Self concept and sense of social inclusion Sing Up children s actudinal quesjonnaire: Overall results (n= 10,425 quesjonnaires) Singing development, self concept and social inclusion CorrelaJon between sense of being social included and singing ability is similar for both groups (n = 6087) Girls more posijve than boys on all themes Younger children tend to be more posijve than older children Sing Up experienced children tend to be significantly more posijve than Non-Sing Up children about singing in school (p=. 001) (Welch et al, 2010; see also 2014 social inclusion arjcle) Normalised Singing Score 85 80 75 70 (Welch, Himonides, Saunders, Papageorgi & Rinta, 2014) 74.62 lowest quarjle 77.94 (p <.0001) But Sing Up parjcipants > Non-Sing Up parjcipants 79.80 low quarjle high quarjle Social Inclusion Quar3le 82.48 highest quarjle 9
Sing Up s Chorister Outreach Programme Primary school children who experienced the Chorister Outreach Programme (n=943) had: the highest posijve mean actudes to singing; the strongest reported engagement with singing; and the highest average social inclusion score And if you do not improve and you find difficulty singing in-tune? (Saunders et al, 2012) Enhancing neurofeedback in singing Real-Jme displays (WinSINGAD) Singing accuracy can be improved through Appropriate pedagogy unambiguous tasks meaningful feedback absence of threat Real-Jme visual feedback [examples] SINGAD/WinSINGAD (Howard et al; Welch et al, 1989: 2005) Sing&See (Thorpe, 2002; Leong & Cheng, 2013) Playing video games (Paney, 2014) (Welch, 1985; Welch et al 2005) 10
Real-Jme displays (Sing&See) Amusic development Group of 5 amusics 7 weekly informal 1½ hour singing workshops including using Sing&See sotware Amusic s song singing showed significant changes pre- to post-intervenjon, parjcularly in singing Happy Birthday (Anderson et al, 2012 - NYAS) Neuropsychobiology of Singing The New London Orchestra s Literacy through Music programme (2010-2011) singing = a distributed neural system in acjon Mean%reading%ages%Start%of%NLO%Programme% Mean%reading%ages%End%of%NLO%Programme% reading 7.5% = examples of wider impact? 6.89% 7.05% 6.85% Control% Interven2on% (Welch et al, 2012; Saunders et al 2014) 11
Evidence of strong links between Music educajon and dyslexia rhythm pulse singing rhythmic speech sound percepjon phonological development reading speaking Study of n=68 adults 22 non-dyslexic musicians 25 dyslexic musicians 21 dyslexic non-musicians Range of psychoacousjc tasks (frequency, durajon, amplitude, rise Jme, fall Jme, ISI intervals between sjmuli) Clear evidence of beneficial impact of musical training on auditory funcjoning For all of the auditory measures, musicians with dyslexia showed equivalent auditory processing to non-dyslexic musicians Dyslexic musicians were significantly more sensijve in most of the auditory tasks than dyslexic controls. Dyslexic musicians also had superior phonological processing skills Based on studies of children s reading development and hearing impairment, as well as child and adult studies of music and dyslexia, music and physical rehabilitajon ater trauma (Bishop-Liebler et al, 2014) (Ford Thompson & Schlaug, 2015, p36/p38 Scien,fic American Mind) The sequence below shows how, over Jme, melodic intonajon therapy (MIT) built up connecjons between the hearing and speaking regions in a healthy right brain. The pajent had suffered a massive stroke at age 11 that had destroyed her speech pathways in the let hemisphere. Over Jme, with therapy, she learned to draw on undamaged brain regions that moderate the rhythmic and tonal aspects of language, bypassing the speech pathways on the let side of her brain that were destroyed. In other words, she found her way back to language through music. Can singing training improve hearing-impaired children s hearing abilities and voice use? N=34,aged 5-7 years, 20-week singing programme, including visual sjmuli and feedback When she began therapy in 2008, she could not string together more than two or three words, and her speech was oten ungrammajcal, leaving her frustrated whenever she tried to communicate. Her treatment plan was intensive an hour and a half a day for up to five days a week, with 75 sessions in all. By the end of the 15-week treatment period, she could speak in sentences of five to eight words, somejmes more. Over the next several years she treated herself at home using the techniques she learned during the sessions. Today, eight years ater her stroke, she spends some of her Jme as a mojvajonal speaker, giving hope and support to fellow stroke survivors. (Welch, Vickers et al 2015) Singing accuracy of pitch and vocal produc,on score (blind scoring at the IoE by trained listeners) HI group far poorer than the NH group from the outset Improvements observed for HI group across sessions 12
N=34 N=34 Singing range Pitch PercepJon Children asked to sing a rising and a Piano-tone three note chord task falling glide and scored on lowest and highest note Three-alterna,ve forced choice task, one s,mulus out of three was different. Pass or fail for each contrast. HI group far poorer than the NH group from the outset HI were poor at the task in the first session (2 children scored above zero), HI improved over,me - in the third Improvements observed for HI session 8 children scored above zero group across sessions NH children were all able to do the task HI became closer to NH over,me and improved over,me; problem with a ceiling effect (Welch, Vickers et al 2015) (Welch, Vickers et al 2015) Conclusions (1) Singing: A conjnuum of vocal ability Singing is a whole brain acjvity Singing development is normal Singing competency develops over Jme and also relates to gender, ethnicity, age, neurotypical status and opportunity What counts as singing is shaped by the socio-cultural experiences Development can be nurtured or hindered by experience Tone deafness (a) is extremely rare in a clinical sense; and (b) suscepjble to pedagogical intervenjon abnormal/ atypical normal neurotypical supranormal Specific learning difficuljes? 13
Conclusions (2) Where a nurturing singing culture is experienced (both at home and elsewhere), children are more likely to gain mastery, and exhibit mastery at an earlier age compared to their less favourably experienced peers singing disability transforms into ability across childhood for the majority In other cases, specific pedagogical intervenjon can make a posijve difference and foster ability Suggested Key Readings on Singing Singing and social inclusion (2014) hjps://www.researchgate.net/publicajon/264793326_singing_and_social_inclusion Using singing to nurture children s hearing (2015) hjps://www.researchgate.net/publicajon/ 281635811_Using_singing_to_nurture_children%27s_hearing_A_pilot_study Singing and Vocal Development (2006); and see revised version 2016 just published by OUP hjps://www.researchgate.net/publicajon/ 279063526_Singing_and_Vocal_Development Singing as communicajon (2005); and see Welch & PreJ (2016) in Oxford Handbooks Online on same topic hjps://www.researchgate.net/publicajon/265579139_singing_as_communicajon Singer idenjjes and educajonal environments (2015) to be published by OUP 2016 hjps://www.researchgate.net/publicajon/ 279956790_Oxford_Handbook_of_Musical_IdenJJes_Chapter_30_Singer_idenJJes_an d_educajonal_environments Welch (2015) Amsterdam Key Readings on Singing (conjnued) AssociaJons between early shared music acjvijes in the home and later child outcomes (2015) hjps://www.researchgate.net/publicajon/ 272174488_AssociaJons_between_early_shared_music_acJviJes_in_the_home_ and_later_child_outcomes_findings_from_the_longitudinal_study_of_australian _Children Sex, gender and singing development (2012) hjps://www.researchgate.net/publicajon/ 245032319_Sex_Gender_and_Singing_Development_Making_a_PosiJve_Differen ce_to_boys%27_singing_through_a_najonal_programme_in_england There will be other publicajons of interest: see the ResearchGate pages for Welch hjps://www.researchgate.net/profile/graham_welch/contribujons; also the Oxford Handbooks online entries where my work is published and/or quoted hjp://www.oxfordhandbooks.com/search? q=welch&searchbtn=search&isquicksearch=true Research is a collaborajve acjvity Special thanks to: Acknowledgements All our parjcipants, especially the children (and, in some cases, their parents and carers) The NaJonal Singing Programme Research Team (Dr Evangelos Himonides, Dr Jo Saunders, Dr Ioulia Papageorgi, Dr Tiija Rinta, Dr Costanza PreJ, Dr Cynthia Benson, Dr Paula Bishop-Liebler, Dr Maria Vraka, Joy Hill, Marc Sarazin) The coordinators for the Sing Up NaJonal Singing Programme, Choir Schools AssociaJon, New London Orchestra, Proge^o Musica, Newham s Every Child a Musician team, European Concert Hall OrganisaJon (ECHO), In Harmony (Opera North; Sage Gateshead) Dr Leon Thurman; Professor Larry Parsons Dr Debi Vickers and the UCL Hearing Unit team Our doctoral and masters students Teachers, headteachers, tutors and administrators Milton, Alexander and Sally for inspirajon g.welch@ioe.ac.uk 14
g.welch@ioe.ac.uk www.imerc.org 15