PronSIG Webinar 2020 Sme Sullivan 1

pronSIG webinar 2020sme sullivan1

We can’t teach pronunciation? Youcan’t teach pronunciation –and that’s the wrong problem In this webinar we’ll focus on the ballet oflanguage sounds, and on the brainnetworks recruited to process speech.pronSIG webinar 2020sme sullivan2

We can’t teach pronunciation?1. THE LANGUAGE2. THE BRAIN3. SPEECHSTREAM a way of learning to hear.pronSIG webinar 2020sme sullivan3

We can’t teach pronunciation?MYTH The written word is somehow superiorto the spoken language. The written code is a measuring stickfor speech, or for correct speech. That we can teach pronunciation fromthe basis of the written word.pronSIG webinar 2020sme sullivan4

We can’t teach pronunciation?Richard Cauldwell. At 4:12 minshttps://youtu.be/QQLLFQ -VHs Listening Decoding in Use 1 May 2020pronSIG webinar 20205sme sullivan

Richard Cauldwell “But learners increasingly aretelling us that they are bamboozled when itcomes to real life listening.”At 4:12 minshttps://youtu.be/QQLLFQ -VHs Listening Decoding in Use 1 May 2020pronSIG webinar 20206sme sullivan

Listening Decoding20132018Richard Cauldwell – www.speechinaction.orgAnd – Listening Decoding. May 2020.Four videos: https://youtu.be/QQLLFQ -VHs

We can’t teach pronunciation?Hickok & Poeppel 20-year Retrospective 12 May 2020https://youtu.be/6GgeLbhXeCgpronSIG webinar 2020sme sullivan8

We can’t teach pronunciation?MYTH Poeppel (at 54 minutes)."bothneuroscientists and linguists are notdoing a great job dealing with that" [thepractical problem of working together,sharing insights] pronSIG webinar 2020 // sme sullivan9

We can’t teach pronunciation?I want to share the informationthat brought me to this positionabout pronunciation,a journey involving aa good hard look at whatreal live language was like.pronSIG webinar 2020sme sullivan10

We can’t teach pronunciation?A journey through research inneuroscience & cognitive sciencewith the eyes of a language teacher.pronSIG webinar 2020sme sullivan11

We can’t teach pronunciation?1. THE LANGUAGE2. THE BRAIN3. SPEECHSTREAM a way of learning to hear.pronSIG webinar 2020sme sullivan12

We can’t teach pronunciation? My adult Asian studentscouldn’t hear fluent English; Their children learnt quickly and well;How could I give them thefluency in their second language (L2)that their children enjoyed?pronSIG webinar 2020sme sullivan13

We can’t teach pronunciation?Was there any way I could open up thosechildhood pathways- thoseautomatic language learning networksin the brain?pronSIG webinar 2020sme sullivan14

languageA journey involving a close look atreal live language.What is it really like?pronSIG webinar 2020sme sullivan15

languageTHE LANGUAGE Rhythm and melody?Hammer words & grammar words;Content & function words;Real live language.pronSIG webinar 2020sme sullivan16

languageTHE LANGUAGE rhythm & melodyWhy did you do that?pronSIG webinar 2020sme sullivan17

languageTHE LANGUAGE rhythm & melodyWhat did you have for dinner?pronSIG webinar 2020sme sullivan18

languageContent and function words:Content wordscarry the meaning (nouns, main verbs, adjectives).Function wordscarry the grammar (auxiliary verbs, prepositions, articles functional relationships between things.pronSIG webinar 2020sme sullivan19

languageWhat is it really like?ther nitsgah nah tern tuhfu much uh thui zi ni . i snomi du vbeen shu di vbeenpronSIG webinar 2020sme sullivan20

languageWhat is it really like?ther nitsgah nah tern tuh shou wizthen it’s gonna turn to showersfu much uh thui zin thu . i snofor much of the countryisn’t the .it’s notmi du vbeen shu di vbeenmight have been should have beenpronSIG webinar 2020sme sullivan21

languagei zin thu . i snoisn’t the .it’s notmi du vbeen shu di vbeenmight have been should have been1. sker na beesim Mor ning klow2. a pula se tar3. a pul hazin sedapronSIG webinar 2020sme sullivan22

languagesker na beesim Mor ning klowa pula se tara pul hazin sedapronSIG webinar 2020sme sullivan23

language sker na beesim Mor ning klow‘s gonna be some morning cloud a pula se tarApple are set to launch a pul hazin seda wordApple hasn’t said a wordpronSIG webinar 2020sme sullivan24

language tha sa nii sweater I zi new? yea zi dizpronSIG webinar 2020sme sullivan25

language tha sa nii sweaterthat’s a nice sweater I zi new?is it new? yea zi dizyes it isSmall Talk – Carolyn Graham (1986)pronSIG webinar 2020sme sullivan26

brainTHE BRAIN two memory networks; uptake to procedural memory; forming a sound system& grammar system.pronSIG webinar 2020sme sullivan27

brainA journey through neuroscience& cognitive science: Two memory networks Uptake of “silent” dataacoustic data from the speech signal Text blocks processingpronSIG webinar 2020sme sullivan28

brainTHE BRAINTwo memory networks(Ullman, 2001)- Procedural for skills- Declarative for factspronSIG webinar 2020sme sullivan29

essesL1 speakersmakemost use ofautomatedimplicitprocesses.30

essesAdultL2 speakersmake mostuse of slowerword by word,explicitprocesses.31

essesThe more fluentthat L2 speakersare, the moreuse of implicitprocesses theymake (Ullman 2001).32

brainProcedural memory:we can’t hearthe uptake is implicitautomaticunstoppableunstartableIt’s automatic FOR CHILDREN.pronSIG webinar 2020sme sullivan33

brainDeclarativenetworkCHILDRENL1,L2 & L3Vocabulary,reading &writingL1 soundandgrammarProcedural networkL2 sound andgrammarL3 sound &grammar34

brainDeclarative networkADULTSL1 & L2Vocabulary,reading & writing.L2 SOUNDS &GRAMMARProcedural networkL1 soundandgrammarL2 sound andgrammar35

brainTHE BRAIN two memory networksHOW DOES SPEECH GET TAKEN UPINTO PROCEDURAL MEMORY?pronSIG webinar 2020sme sullivan36

brain How does speech gettaken up into procedural memory? What is “under the bonnet” inthe brain? Which language features aretaken up into procedural networks? Why can’t we use all that for adults?pronSIG webinar 2020sme sullivan37

brainMichel Paradis (2009)Studies in Bilingualism, 40. John Benjamins.pronSIG webinar 2020sme sullivan38

brainWe acquire the first language from surface statistics of the speech signal,e.g. Formants & Wavebursts.pronSIG webinar 2020sme sullivan

owel.htmlpronSIG webinar 2020sme sullivan40

brainWhat goes on “under the bonnet”?The establishment of a system of weighted connectionsinduced by tallying the frequency of occurrence of anycombination of items.Both what is tallied and the tallying process are implicit.We don’t know they’re happening and we can’tinfluence the process.pronSIG webinar 2020sme sullivan41

brainDownload APP Brain Anatomy http://play.google.com/store/apps/details?id com.visual3dscience.brainpronSIG webinar 2020sme sullivan42

brainprocedural memoryactivatescerebellum,basal ganglia&perisylvian cortical areasdeclarative memoryactivateshippocampal system,parahippocampal gyri,mesial temporal cortex& anterior cingulatepronSIG webinar 2020sme sullivan43

brainpronSIG webinar 2020sme sullivan44

brainWhat has this got to do with pronunciation? “axonal rewiring” (Uylings 2006) for these functions whichhave a critical period like language may be difficult. Specially designed strategies might be necessary to obtain betterplastic changes in adult learning. MacWhinney, Squire, Uylings and others in neurocognitive fieldsuggest a rich environment of repeated input may induceplasticity in adult networks.45

brainWhat has this got to do with pronunciation? Decoding fluent speech requires automatic linguisticcompetence in segmenting, in phonologicalrepresentations and in syntactic competency. The linguistic demands of segmenting fluent speechand accessing adequate automatic phonologicalrepresentations requires fast procedural memorynetworks. Automatic linguistic procedures are inadequatelyinitiated and activated in AL246

SPEECHSTREAMSPEECHSTREAMpronSIG webinar 2020sme sullivan47

.speechstreamSpeech stream exercisesAn experience where students Listen and repeat fluent natural speech Which is partly incomprehensible to them Which they will gradually unfold/ decodewith teacher support.48

speechstreamThe speechstream exercise The decoding process happens when understanding is absent. 49When understanding is absent,processing occurs.

speechstreamSpeechstreamI wanted to trick the childhood learning networksinto opening up for my adult students.I concluded from my reading & experience thatTEXT blocks adult uptake because textactivates their first language sound system.50

brainADULTSreading &writingL1 soundandgrammarL2 sounds &grammarL2 sound andgrammarReading & writing alwaysactivate a sound system.We always sound the words we read,even if we don’t notice that.51

speechstreamTEXT blocks adult uptakeBecause text activates their first languagesound system, the brain cannot create anew sound system for adults if the old one isactive.Paradis, (2009)52

SPEECHSTREAM No text; No understanding; Intense repetition; No homework; No re-visiting the bunch of words; Encouragement & trust; Enjoyment.pronSIG webinar 2020sme sullivan53

SpeechstreamSTAGE 1:scanningSTAGE 2:decodingPROCEDURAL MEMORYDECLARATIVE MEMORYca. 5 – 10 minsca. 10 to15 minutesSCANNING . DECODING intense focussed repetition . begin finding words NOT understanding.continue articulating physical effort. various exercises Feedback54with sound, grammar

speechstreamWhen the brain focuses onincomprehensible speech it opens upthe processor to deal with it . Andbegins creating a second automaticsound system for the new language.pronSIG webinar 2020sme sullivan

speechstreamStudents enjoy speechstreamdespite the effort involved –because they appreciate theimmediate delight of decoding thetangle – and the longer termprogress in better listening to reallive language.pronSIG webinar 2020sme sullivan

We can’t teach pronunciation, can we?We can’t teach pronunciation,can we? YES!We will continue to use therich resources we have available.And we can take the information that isoverflowing from the neuroscience fieldlooking at new approaches and new anglesfor creating the best learning environmentfor adult learners.pronSIG webinar 2020sme sullivan57

languageMany thanks to pronSIG, Beata, Adam& Gemma andto all the participants.Hope you enjoyed our webinar.You can contact me atspeechstreamsue@gmail.compronSIG webinar 2020sme sullivan58

PRONUNCIATIONReferencesCauldwell, R. (2014). Listening and pronunciation need separate models of speech. In J.Levis & S. McCrocklin (Eds). Proceedings of the 5th Pronunciation in Second LanguageLearning Teaching Conference (pp.). Ames, IA: Iowa State University.Paradis, M. (2009). Declarative and Procedural Determinants of Second Languages, JohnBenjamins Pub Co.Pierrehumbert, J. B. (2003). "Phonetic diversity, statistical learning, and acquisition ofphonology." Language and speech 46(2-3): 115-154.Poeppel, D. (2014, November 18). Dr. David Poeppel, New York University [Videofile].Retrieved https://www.youtube.com/watch?v lIqO4wz3VCsSquire, L. R., Ed. (2003). Fundamental Neuroscience, Academic Press.Ullman, M. T. (2001). "A neurocognitive perspective on language: Thedeclarative/procedural model." Nature Reviews Neuroscience 2(10): 717-726.Ullman, M. T. (2004). "Contributions of memory circuits to language: thedeclarative/procedural model." Cognition 92(1-2): 231-270.pronSIG webinar 2020sme sullivan

a way of learning to hear. We can’t teach pronunciation? pronSIG webinar 2020 sme sullivan 4 MYTH The written word is somehow superior to the spoken language. The written code is a measuring .