Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Speech errors and segment duration : an investigation of word-initial/sp, st, sk/-clusters under conditions… Pyplacz, Verna 1976

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1976_A6_7 P96.pdf [ 4.55MB ]
Metadata
JSON: 831-1.0093697.json
JSON-LD: 831-1.0093697-ld.json
RDF/XML (Pretty): 831-1.0093697-rdf.xml
RDF/JSON: 831-1.0093697-rdf.json
Turtle: 831-1.0093697-turtle.txt
N-Triples: 831-1.0093697-rdf-ntriples.txt
Original Record: 831-1.0093697-source.json
Full Text
831-1.0093697-fulltext.txt
Citation
831-1.0093697.ris

Full Text

SPEECH ERRORS AND SEGMENT DURATION: AN INVESTIGATION OF WORD-INITIAL /sp, s t , sk/-CLUSTERS UNDER CONDITIONS OF RAPID REPETITION by Verna Lynne Pyplacz B.A., U n i v e r s i t y of B r i t i s h Columbia, 1974 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of P a e d i a t r i c s D i v i s i o n of Audiology and Speech Sciences We accept t h i s t h e s i s as conforming to the r e q u i r e d standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1976 (c) Verna Lynne Pyplacz, 1976 In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th is thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thes is fo r f i nanc ia l gain sha l l not be allowed without my wr i t ten permission. Verna Lynne Pyplacz Department of P a e d i a t r i c s  D i v i s i o n of Audiology and Speech Sciences The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date 14th May 1976. ABSTRACT Speech e r r o r s , or " s l i p s of the tongue", have been studied i n attempts to understand the speech production process, to i n v e s t i g a t e p h o n o l o g i c a l u n i t s and r u l e s , and to provide i n s i g h t s i n t o h i s t o r i c a l l i n g u i s t i c change. The present study examines speech e r r o r s and t h e i r r e l a -t i o n to segment durations i n w o r d - i n i t i a l /sp, s t , sk/-c l u s t e r s produced under r a p i d r e p e t i t i o n c o n d i t i o n s by s i x adult n a t i v e speakers of E n g l i s h . F i f t y percent of the e r r o r s produced could be c l a s s i -f i e d as r e p e t i t i o n e r r o r s ; these were examined f o r dur a t i o n i n the i n i t i a l c l u s t e r s , both e r r o r and c o r r e c t e d productions. General r e s u l t s f o l l o w i n g from a n a l y s i s of the data were: (1) E r r o r c l u s t e r s and t h e i r component segments were con-s i s t e n t l y longer i n du r a t i o n than t h e i r subsequent and imme-di a t e c o r r e c t i o n s . (2) The c l u s t e r s /sp/ and /sk/ are longer than / s t / , which may be a t t r i b u t a b l e to the f a s t e r moving, more h i g h l y i n n e r -vated tongue t i p musculature i n v o l v e d i n the production of Is/ and It/, compared w i t h the heterorganic c l u s t e r s . (3) The stop consonant i n a given c l u s t e r appears to deter-mine the o v e r a l l c l u s t e r d u r a t i o n , since the d u r a t i o n of /s/ remains f a i r l y constant i r r e s p e c t i v e of context. In l i g h t of the r e s u l t s , i t was speculated that the ex-ces s i v e d u r a t i o n of the c l u s t e r (or of i t s component p a r t s ) v i o l a t e d a timing c o n s t r a i n t on the production of an u t t e r -ance, n e c e s s i t a t i n g r e c a l i b r a t i o n and c o r r e c t i o n of the e r r o r . I t was f u r t h e r i n f e r r e d that feedback must be present i n order f o r the system to recognize the d u r a t i o n e r r o r , to compare i t w i t h planned output, and f i n a l l y to execute a c o r r e c t i o n . Two types of feedback were considered necessary f o r the adequate f u n c t i o n i n g of a speech production model, which would a l s o a l l o w f o r speech perc e p t i o n : (a) continuous a u d i t o r y feedback, which i s supplemented by (b) i n t e r m i t t e n t p r o p r i o c e p t i v e feedback, both of which are used i n p e r c e i v i n g input ,:and manipulating output. Such a system provides a p l a u s i b l e account of speech e r r o r production as described i n t h i s study. The hypothesized v a r i a b l e servomonitor system advocated here (and i n other s t u d i e s ) i n general provides an e f f i c i e n t means f o r producing, monitoring and c o r r e c t i n g speech production. TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i LIST OF FIGURES v i i i ACKNOWLEDGEMENTS i x DEDICATION x Chapter 1 INTRODUCTION 1 1.0 I n t r o d u c t i o n 1 1.1 L i t e r a t u r e Review: I n t r o d u c t i o n 3 1.2 F u n c t i o n a l Neuroanatomy of Speech 3 1.21 C e n t r a l Nervous System 4 1.22 P e r i p h e r a l Nervous System 9 1.23 Sensory T r a c t s H 1.24 Motor Systems .; 12 1.3 Feedback Mechanisms :. 13 1.30 I n t r o d u c t i o n 13 1.31 The Gamma / Spin d l e Motor System 14 1.32 A u d i t o r y Feedback 19 1.33 Summary 21 1.4 Speech E r r o r s 22 1.5 Timing of Speech 25 1.6 Models of Speech Production 26 1.60 I n t r o d u c t i o n 26 1.61 Open-Loop Models 27 1.62 Closed-Loop Models 28 1.63 Summary and D i s c u s s i o n 30 i v Page Chapter 2 STATEMENT OF THE PROBLEM 34 Chapter 3 METHOD 36 3.1 P i l o t Studies 36 3.11 P i l o t Study I 36 3.12 P i l o t Study II 37 3.13 D i s c u s s i o n 38 3.2 Main Study ; . . 39 3.3 A n a l y s i s of Data .' 40 Chapter 4 RESULTS 44 4.0 I n t r o d u c t i o n 44 4.1 C o n t r o l Group Data 47 4.2 Experimental Group Data 54 Chapter 5 DISCUSSION 64 5.0 General Co n s i d e r a t i o n s 64 5.1 D i s c u s s i o n of the C o n t r o l Group 65 5.2 D i s c u s s i o n of the Experimental Groups 67 5.3 T h e o r e t i c a l C o n s i d e r a t i o n s 71 5.4 L i m i t a t i o n s of the Present Study 76 5.5 Summary and Conclusions 77 REFERENCES 80 APPENDIX A 85 APPENDIX B 86 APPENDIX C 87 APPENDIX D 88 v LIST OF TABLES Table Page 1 Types and Numbers of Speech E r r o r s Produced by Each Subject 46 2 Types and Numbers of R e p e t i t i o n E r r o r s Produced by Each Subject 46 3 C o n t r o l Group Data (Normal Co n v e r s a t i o n a l Rate) f o r W o r d - I n i t i a l Consonant C l u s t e r s /sp-/, / s t - / , /sk-/: Means and Standard D e v i a t i o n s of Segmental Durations ( i n ms) 48 4 Summary of A n a l y s i s of Variance: C o n t r o l Group Data -- C l u s t e r Segments and Subjects = .. 50 4a Newman-Keuls Summary Table: C o n t r o l Group Data --C l u s t e r Segments and Subjects 50 5 Summary of A n a l y s i s of Variance: C o n t r o l Group Data -- C l u s t e r s and Subjects . 51 5a Newman-Keuls Summary Table: C o n t r o l Group Data --C l u s t e r s and Subjects 51 6 Summary of A n a l y s i s of Variance: C o n t r o l Group Data -- C l u s t e r Segments, One-Way C l a s s i f i c a t i o n 52 6a Newman-Keuls Summary Table: C o n t r o l Group Data --C l u s t e r Segments 52 7 Summary of A n a l y s i s of Variance: C o n t r o l Group Data --- C l u s t e r s , One-Way C l a s s i f i c a t i o n 53 7a Newman-Keuls Summary Table: C o n t r o l Group Data C l u s t e r s 53 v i Table Page 8 Summary S t a t i s t i c s f o r Segmental and C l u s t e r Durations ( i n ms), i n S i x R e p e t i t i o n E r r o r Groups (termed Experimental Groups) 56 9 Summary of R e p e t i t i o n E r r o r Data f o r Experimental Groups #2-#5: Segmental and C l u s t e r Durations ( i n ms) ; . 61 v i i LIST OF FIGURES Figur e Page 1 Block Diagram of Instrumentation Used to Produce Oscillograms 42 2 C o n t r o l Group Data: Segment and C l u s t e r Duration; Mean values f o r each subject 55 3 Experimental Groups #2-#5: Segment and C l u s t e r Duration; Mean values f o r e r r o r and c o r r e c t productions . 59 4 Experimental Groups #1 and #6: Segment and C l u s t e r Duration; Mean values f o r e r r o r and c o r r e c t productions 60 v i i i ACKNOWLEDGEMENTS I wish to express my g r a t i t u d e to a l l those who gave of t h e i r time and energy i n h e l p i n g me complete t h i s t h e s i s . My thanks go e s p e c i a l l y t o : — Dr. John G i l b e r t -- f o r h i s time and patience given as my su p e r v i s o r ; — Dr. Dale Kinkade -- f o r t r a n s c r i b i n g data, g i v i n g c r i t i c a l comments, and being k i n d enough to serve on my committee; — Dr. John Delack -- who doesn't s u f f e r f o o l s g l a d l y , f o r h i s time, e f f o r t and necessary c r i t i c a l comments; — Ronnie S i z t o -- f o r days (and n i g h t s ) spent p a t i e n t l y running, debugging, rerunning and i n t e r -p r e t i n g s t a t i s t i c a l programmes at the eleventh hour; — Dr. Brenda F r a z e r - - f o r her s t a t i s t i c a l advice; — I n g r i d J e f f r e y -- f o r her advice on a n a l y z i n g o s c i l l o g r a m s ; — My Subjects -- f o r t h e i r p atience i n rep e a t i n g "tongue-t w i s t e r s " ; — My Family -- who are always there when I need them; — My Friends i n the Graduating C l a s s of '76. i x This t h e s i s i s dedicated to the memory of my f a t h e r , Stan Pyplacz ( 1 9 2 2 - 1 9 7 5 ) . x CHAPTER 1 INTRODUCTION 1. 0 I n t r o d u c t i o n A speech e r r o r can be, and has been, v a r i o u s l y described as a "spoonerism" ( a f t e r the Revd. W i l l i a m Spooner), a "port-manteau" word (coined by Lewis C a r r o l l ) or a " s l i p of the tongue" and can best be defined as "an u n i n t e n t i o n a l l i n g u i s -t i c i n n o v a t i o n " (Sturtevant,, 1947, p. 38). Speech e r r o r s are con s t r a i n e d by the grammar and phonology of a given language, and because of these c o n s t r a i n t s they are to a c e r t a i n degree p r e d i c t a b l e and non-random (Fromkin, 1973, p. 113). Authors such as Shakespeare, Rabelais and Lewis C a r r o l l used speech e r r o r s i n t h e i r works to achieve humourous ends. Freud b e l i e v e d that these disturbances of speech were "the r e s u l t of complicated p s y c h i c a l i n f l u e n c e s , of elements outside the same word, sentence or sequence of spoken words" (Freud, 1924; c i t e d i n Fromkin, 1973, p. 110). Speech e r r o r s have also been s t u d i e d i n the hope that such would provide i n s i g h t i n t o h i s t o r i c a l l i n g u i s t i c change (Stur t e v a n t , 1947), i n attempts to understand the speech production process more f u l l y (Boomer 5 Laver, 1968; Fromkin, 1968, 1971; Nooteboom, 1969) , and to i n v e s t i g a t e the p o s s i b l e bases f o r c e r t a i n p h o n o l o g i c a l u n i t s and r u l e s (Fromkin, 1968, 1971). 1 2 Few i n v e s t i g a t o r s have speculated as to the underlying cause of speech e r r o r s , although Meringer (1908) t r i e d --and f a i l e d -- to c o r r e l a t e e r r o r production w i t h numerous v a r i a b l e s , such as rate of speech and time of day. While speech e r r o r s have been used as a v e h i c l e f o r i n v e s t i g a t i n g various speech processes, t h i s temporary "breakdown and re-c a l i b r a t i o n " process i n which the system i s i n v o l v e d can be s t u d i e d i n i t s own r i g h t ; i . e . , an examination of the i n -t r i n s i c s t r u c t u r e of speech e r r o r s i s l o g i c a l l y p r i o r to t h e i r use f o r other purposes. The methodologies employed to c o l l e c t speech e r r o r data, as w e l l as the subsequent c l a s s i f i c a t i o n s of such data, have been many and d i v e r s e . The speech e r r o r s c o l l e c t e d i n the present study, f o r example, could be c l a s s i f i e d and accounted f o r almost e n t i r e l y by the d e s c r i p t i o n s provided i n Fairbanks and Guttman (1958) ; w i t h regard s p e c i f i c a l l y to the " r e p e t i -t i o n e r r o r s " found i n the present study, researchers i n the f i e l d of delayed auditory feedback (DAF) have encountered a s i m i l a r phenomenon which they have l a b e l l e d " a r t i f i c i a l s t u t t e r " ( c f . Lee, 1951). In the DAF l i t e r a t u r e , i t has been noted that subjects r e q u i r e a " t u r n around time" ( i . e . , a delay i n which a subject can produce a r e p e t i t i o n a f t e r a f i r s t p r o d u c t i o n ) , and t h i s delay time could have a neuro-p h y s i o l o g i c a l b a s i s , such as that proposed by Kent and M o l l (1975). 3 An examination of the l i t e r a t u r e p e r t a i n i n g to func-t i o n a l neuroanatomy and the neurophysiology of motor res-ponses (e.g., Bowman, 1971; Abbs, 1973) suggests that t h i s delay a r i s e s as a r e s u l t of c o r t i c a l l y c o n t r o l l e d motor mechanisms (gamma and alpha motor systems); i . e . , delay time might be accounted f o r by means of a gamma "delay" loop f o r feedback from the p o s i t i o n and movement of the a r t i c u l a t o r s (tongue, jaw, etc.) during speech. 1.1 Review of the L i t e r a t u r e - I n t r o d u c t i o n The l i t e r a t u r e r e l e v a n t to t h i s study w i l l be discussed . i n four s e c t i o n s : (1) a-.i o u t l i n e of the f u n c t i o n a l neuro-anatomy of speech, (2) feedback mechanisms, (3) speech e r r o r s , and (4) timing of speech. The f i f t h s e c t i o n w i l l provide a d i s c u s s i o n and summary of models of speech production i n an attempt to synthe s i z e i n f o r m a t i o n from the above-mentioned d i s c i p l i n e s . 1.2 Fu n c t i o n a l Neuroanatomy of Speech As has often been noted (e.g., MacNeilage, 1972, pp. 6-7), the importance of p h y s i o l o g i c a l mechanisms f o r speech i s t h e i r i n t e r a c t i o n i n the production of an a c o u s t i c out-put which has communicative s i g n i f i c a n c e . In t h i s s e c t i o n a b r i e f overview of some of the main areas of the b r a i n w i t h s p e c i f i c s i g n i f i c a n c e i n generating speech w i l l be discussed, as w e l l as the c e n t r a l nervous system, the p e r i p h e r a l nervous system and the sensory and motor t r a c t s , a l l of which make 4 up the pathways f o r speech. The d i s c u s s i o n w i l l descend, anatomically speaking, from the cortex to the thalamus and on downward through the midbrain, pons, medulla, cerebellum, c r a n i a l nerves and s p i n a l cord. 1.21 C e n t r a l Nervous System The c e r e b r a l cortex i s to be regarded as the supreme manipulator of motor neuron impulses r e s u l t i n g ' i n speech. This idea has been debated f o r s e v e r a l decades, culminating w i t h the n o t i o n that the c e n t r a l nervous system (CNS) can be regarded "as a s e r i e s of f u n c t i o n a l arcs i n which sub-c o r t i c a l centers are i n a r e c i p r o c a l r e l a t i o n s h i p with c o r t i -c a l areas" (Berry § Eisenson, 1956, p. 45; c f . also P e n f i e l d § Roberts, 1959, p. 15). These arcs are claimed to i n t e r a c t w i t h one another and not e x c l u s i v e l y w i t h the cortex. This view of the CNS as an input-output r e f l e x arc' i s now, how-ever, somewhat outmoded: Pribram (1971) has described the presence of feedback and feedforward mechanisms of the CNS which c o n t r o l receptor f u n c t i o n s and has hypothesized a f u n c t i o n a l " T e s t - O p e r a te-Test-Exit" servomechanism, which matches input against the output t a r g e t . Whichever system i s at work i n the CNS, the r e s u l t i s a complex i n t e r a c t i v e process: speech. Auditory p e r c e p t i o n , which i s used f o r l e a r n i n g and maintaining speech, i s b e l i e v e d to be found i n the auditory reception area (Brodmann's Area 22). I t i s w i t h i n t h i s area 5 that an i n d i v i d u a l may per c e i v e sounds but not decode t h e i r meaning, t h i s l a t t e r f u n c t i o n being accomplished i n Wer-nicke's area (Area 41-42): "In the auditosensory area [Area 22] auditory impressions reach consciousness as sounds, and t h e i r loudness, q u a l i t y and p i t c h can be d i f f e r -e n t i a t e d . The d i r e c t i o n from which the sound comes and i t s c h a r a c t e r , whether rhythmical or ar h y t h m i c a l , are a l s o determined by t h i s p a r t of the cortex. The s i g n i f i c a n c e and the source of the sound, however, r e q u i r e the a d j o i n i n g audito-psychic area f o r t h e i r e l u c i d a t i o n .... In t h i s area [Areas 41-42] a u d i t o r y impressions r e c e i v e t h e i r i n t e r p r e t a t i o n and can be d i f f e r e n t i a t e d from one another, as regards t h e i r probable source and o r i g i n , by a s s o c i a t i o n w i t h past experience." (Johnston £ W h i l l i s , 1954, p. 1037) I t has been demonstrated i n recent years that loudness, qual-i t y and p i t c h can also be d i f f e r e n t i a t e d s u b - c o r t i c a l l y . In the p a r i e t a l lobe are Areas 1-3 which make p o s s i b l e awareness of touch, pressure, temperature and muscle movement. An awareness of tongue movements i n a r t i c u l a t i o n may be pro-j e c t e d from here to speech areas and "may be one of the c h i e f s t i m u l i i n provoking or c o n t i n u i n g speech" (Berry § Eisenson, 1956, p. 56). The motor p r o j e c t i o n area f o r voluntary movement (Area 4) has a lar g e percentage devoted to phonatory and a r t i c u l a t o r y movement which sends impulses v i a t h i s pyramidal t r a c t to the 6 muscles of the jaw, l i p s , tongue, larynx and pharynx. Area 6, the extrapyramidal area, produces refinement i n motor behaviour such as the sequencing of vocal f o l d adduc-t i o n , resonance and a r t i c u l a t i o n , or the q u a l i t i e s of intona-t i o n and rhythm. Area 44, Broca's area, i s where f i b r e s from other areas concerned w i t h the speech process synapse and then proceed to the motor p r o j e c t i o n areas f o r the muscles of speech. Other areas which may be s i m i l a r i n f u n c t i o n to Area 44 are Areas 7A, 7B and 7C, the l a s t of which i s concerned w i t h the thalamus and w i t h i n t e g r a t i o n of emotional expression i n t o speech. Areas 8-11, the f r o n t a l i d e a t i o n a l a s s o c i a t i o n areas, are c a l l e d upon to i n t e g r a t e past experience,, ab s t r a c t t h i n k i n g , reasoning and ideas i n t o speech. The s t r i a t e bodies, comprised of the caudate, l e n t i c u l a r and amygdaloid n u c l e i , together w i t h the cortex and the thalamus, probably act as one u n i t or arc ( c f . P e n f i e l d § Rasmussen, 1950, pp. 106-107). The caudate and l e n t i c u l a r n u c l e i belong to the extrapyramidal system, and t h e i r axons run to motor n u c l e i of the b r a i n stem concerned with innerva-t i o n of the muscles of the tongue, face, larynx and pharynx. The highest i n t e g r a t i v e mechanism f o r speech may, as P e n f i e l d and Rasmussen (1950, p. 219) suggest, "be s i t u a t e d 7 i n some c e r e b r a l area, such as the thalamus, and not i n e i t h e r c e r e b r a l cortex". A l l sensory t r a c t s have a r e l a y s t a t i o n i n the thalamus. Sensory-emotional responses, the q u a l i t y of the v o i c e , f a c i a l expression and s u b t l e body gestures, as w e l l as conceptual patterns of form, s i z e , q u a l i t y , i n t e n s i t y and t e x t u r e , are organized here f o r trans-mission to the cortex. The midbrain contains the s u b s t a n t i a n i g r a and red n u c l e i , which are p a r t of the extrapyramidal system, and the c e r e b r a l peduncles, which contain the pyramidal t r a c t s . The pyramidal t r a c t i s made up of the c o r t i c o s p i n a l f i b r e s which run from the motor cortex to the s p i n a l cord. This t r a c t has c o n t r o l over the speech muscles of the head and neck through c r a n i a l nerves V, V I I , IX, X and X I I . The extrapyramidal system i s c h i e f l y made up of the s t r u c t u r e s other than the cortex which send impulses to the s p i n a l cord, i . e . , s t r i a t e bodies, cerebellum, red nucleus, s u b s t a n t i a n i g r a , etc. (cf. N e t t e r , 1974). The s u b s t a n t i a n i g r a and the red n u c l e i have two-way connections w i t h the s t r i a t e bodies, thalamus and premotor area of the cortex. The s u b s t a n t i a n i g r a i s b e l i e v e d to c o n t r o l the muscles of f a c i a l e x pression, and the red n u c l e i i n a s s o c i a t i o n with the cerebellum c o n t r o l the gradation and timing of muscular c o n t r a c t i o n . The pons, l o c a t e d j u s t below the midbrain, contains sen-sory and motor pathways, as w e l l as the r e t i c u l a r formation, 8 which i s l i n k e d w i t h the cerebellum and s t r i a t e bodies, making up p a r t of the extrapyramidal system. The pneumotaxic centre of the pons i s connected to the hypothalamus and stim-u l a t e s e x h a l a t i o n and maintains r e s p i r a t o r y rhythm for speech. The t r i g e m i n a l sensory complex i s the p r i n c i p l e sensory nu-cleus of the t r i g e m i n a l (Vth) nerve i n the pons. The motor n u c l e i of the f a c i a l and t r i g e m i n a l nerves i n the pons inner-vate voluntary f a c i a l speech musculature and muscles of masti-c a t i o n , r e s p e c t i v e l y . The medulla contains the centres which c o n t r o l the res-p i r a t o r y and c i r c u l a t o r y systems and also regulate rate and rhythm of b r e a t h i n g f o r speech. These centres respond to incoming sensory impulses from the diaphragm and from the a o r t i c and c a r o t i d c a p i l l a r i e s . The lower motor neurons of the medulla innervate muscles of the mouth, pharynx and larynx f o r speech production v i a s p e c i f i c c r a n i a l nerves (to be discussed below). The medulla contains the nucleus s o l i -t a r i u s which receives a f f e r e n t (sensory) f i b r e s from the f a -c i a l , vagus, and glossopharyngeal nerves. The hypoglossal nucleus of the medulla s u p p l i e s i n n e r v a t i o n to tongue muscles. The nucleus ambiguus sends f i b r e s through the glossopharyn-g e a l , vagus, and s p i n a l accessory nerves to supply muscles of the pharynx and l a r y n x . The cerebellum,part of the extrapyramidal system, i s --i n a d d i t i o n to p r o v i d i n g f o r f i n e motor c o o r d i n a t i o n i n general -- of importance i n speech pr o d u c t i o n , by e l a b o r a t e l y 9 c o n t r o l l i n g voluntary muscle movements, e.g., i n the modula-t i o n of phonation and a r t i c u l a t i o n . The s p i n a l cord conducts sensory impulses to higher c e n t r e s , such as the cerebellum and the thalamus. I t also mediates c o n t r o l of motor a c t i v i t i e s of the body below the face and neck (e.g., posture, movements and gesture). The s p i n a l cord acts as an i n t e g r a t i n g centre f o r many r e f l e x p a t t e r n s . 1.22 P e r i p h e r a l Nervous System The c r a n i a l nerves d i r e c t l y a s s ociated w i t h speech mechanisms are the t r i g e m i n a l (V), f a c i a l ( V I I ) , glosso-pharyngeal ( I X ) , vagus (X), accessory (XI) and hypoglossal ( X I I ) . The t r i g e m i n a l nerve (V), c o n t a i n i n g both sensory and motor f i b r e s important to the a r t i c u l a t o r y movements of speech, transmits sensations of movement from the muscles of m a s t i c a t i o n of the jaw, sensations of touch, temperature and pain from the face, and volun t a r y motor impulses to the jaw.' The f a c i a l nerve ( V I I ) , as w e l l as c r a n i a l nerves I X - X I I , has motor f i b r e s i n n e r v a t i n g the muscles of speech production mechanisms.. The f a c i a l nerve i t s e l f s u p p l i e s the s t r i a t e d muscles of the fa c e , the s t y l o h y o i d muscle and the stapedius muscles. 10 The glossopharyngeal nerve (IX) innervates the s t y l o -pharyngeus muscle, which aids i n v e l a r c l o s u r e . I t l i k e w i s e mediates p r o p r i o c e p t i o n of the p o s t e r i o r t h i r d of the tongue. The vagus nerve (X), i n conjunction with c r a n i a l nerves IX and XI, innervates the voluntary muscles of the pharynx and l a r y n x i n v o l v e d i n speech. Sensory impulses, t r a n s m i t t i n g p r o p r i o c e p t i v e i n f o r m a t i o n to the medulla, cerebellum and other parts of the extrapyramidal t r a c t , can e f f e c t f i n e co-o r d i n a t i o n , graded c o n t r a c t i o n and t o n i c c o n t r o l necessary f o r speech. The s p i n a l accessory nerve (XI) a s s i s t s the vagus i n motor c o n t r o l of the pharynx and larynx and innervates the t r a p e z i u s and sternomastoid muscles necessary f o r speech b r e a t h i n g . C r a n i a l nerve X I I , the h y p o g l o s s a l , innervates the s t r i -ated muscles of the tongue. These muscles concerned w i t h voluntary movements of the tongue are the genioglossus, geni-ohyoid, hyoglossus, s t y l o h y o i d , s t y l o g l o s s u s , and p a l a t o g l o s -sus . There are t h i r t y - o n e p a i r s of s p i n a l nerves, which transmit both sensory and motor information. The most impor-tant r o l e of these nerves i n speech production c o n s i s t s of sending motor ( e f f e r e n t ) impulses to a c t i v a t e the muscles of breat h i n g f o r speech. 11 1.23 Sensory Tracts I t i s of importance to t h i s d i s c u s s i o n to remember that p r o p r i o c e p t i o n (muscle p o s i t i o n , tension and movement) i s r e s p o n s i b l e f o r the r a t e , f o r c e , d i r e c t i o n and extent of v o l u n t a r y movements. One of the l a r g e s t systems of sensory t r a c t s t r a n s m i t t i n g such i n f o r m a t i o n c o n s i s t s of the v e n t r a l and l a t e r a l spinothalamic t r a c t s . As an example of r e l a y s along a sensory t r a c t , we w i l l u t i l i z e the a u d i t o r y pathway. The stimulus i s transformed i n t o an e l e c t r o c h e m i c a l impulse i n the cochlea, and the f i r s t synapse i s l o c a t e d i n the cochlear nucleus of the medulla. The f i b r e t r a c t ascends i n the l a t e r a l lemniscus through the pons to the midbrain. There i s another synapse i n the i n f e r -i o r c o l l i c u l u s and a f i n a l one i n the medial g e n i c u l a t e body of the thalamus, from which the information then passes to the auditory r e c e p t i o n area (Area 22) and to other areas i n the cortex. While the a u d i t o r y impulse i s being t r a n s m i t t e d , p r o p r i -oceptive impulses from muscles i n the tongue, l i p s and jaw (among others) are r e p o r t i n g s h i f t s i n p o s i t i o n and tension which w i l l be used i n the production of speech ( c f . Berry § Eisenson, 1956). I t i s worth n o t i n g here the d i f f e r e n c e s between e f f e r e n t and a f f e r e n t t r a c t s . An a f f e r e n t , or sensory, t r a c t conducts an impulse from the periphery towards the c o r t e x , while an 12 e f f e r e n t , or motor, t r a c t conducts impulses from the CNS to e f f e r e n t nerve endings i n muscles. 1.24 Motor Systems Two t r a c t s comprising the pyramidal system, the c o r t i -c o s p i n a l and c o r t i c o b u l b a r t r a c t s , o r i g i n a t e i n the motor and premotor areas of the cortex. The c o r t i c o b u l b a r t r a c t i s important f o r speech pro-d u c t i o n , since i t a c t i v a t e s the muscles of the tongue, l i p s , jaw, pharynx and lar y n x . The c o r t i c o s p i n a l t r a c t , having c e l l bodies i n the pre-c e n t r a l gyrus ( i . e . , i n the f r o n t a l lobe) of the c o r t e x , makes i t s way v i a the i n t e r n a l capsule (a f i b r e t r a c t ) to the cere-b r a l peduncles ( i . e . , a c r o s s i n g of s e v e r a l f i b r e t r a c t s ) i n the midbrain. This f i b r e t r a c t decussates, or crosses, i n the medulla and then passes i n t o the s p i n a l cord, synapsing w i t h motor c e l l s e f f e c t i n g v o l u n t a r y muscle c o n t r a c t i o n . The c o r t i c o b u l b a r t r a c t f o l l o w s the same route u n t i l i t passes i n t o the pons and medulla, synapsing with lower motor neurons of c r a n i a l nerves V and V I I - X I I . From here f i b r e s of t h i s t r a c t continue on to innervate the muscles of speech. The extrapyramidal system i s also v i t a l to the f i n e l y coordinated motor a c t i v i t y necessary f o r speech. I t s organi-z a t i o n i s described by Grinker and Bucy (1949, p. 274) as f o l l o w s : 13 " A l l of these s u b c o r t i c a l s t r u c t u r e s which are r e c i p i e n t s . o f impulses from the p r e c e n t r a l cor-tex have two p r i n c i p a l p r o j e c t i o n systems. 1) They p r o j e c t to the l a t e r a l nucleus of the thalamus and thence back to the p r e c e n t r a l re-gion (cortex) and 2) they have a descending pathway down the s p i n a l cord to the a n t e r i o r horn c e l l s . ... [The n e c e s s i t y of t h i s system f o r speech a c t i v i t i e s i s that i t ] controls', a c t i v a t e s , and i n h i b i t s the as s o c i a t e d muscula-ture or p r o t a g o n i s t i c muscles which must be appro-p r i a t e l y contracted. ... I t c o n t r o l s the r e f l e x i n n e r v a t i o n of the s k e l e t a l muscles to produce what i s commonly known as tone." [ C i t e d i n Berry § Eisenson, 1956, p. 71] As can be deduced from the above d i s c u s s i o n of neuro-anatomical s t r u c t u r e s and t h e i r f u n c t i o n s , speech can be re-garded as a very complex i n t e g r a t i v e process, encompassing not only motor systems f o r i t s p r o d u c t i o n , but also sensory monitoring systems, which are to be discussed i n more d e t a i l i n the f o l l o w i n g s e c t i o n . 1.3 Feedback Mechanisms 1.30 I n t r o d u c t i o n The goal of t h i s s e c t i o n i s an a p p r e c i a t i o n of the neuro-p h y s i o l o g i c a l b a s i s f o r p e r i p h e r a l p r o p r i o c e p t i v e / k i n e s t h e t i c feedback systems i n ope r a t i o n during the speech act. The pre-s e n t a t i o n here i s germane to the d i s c u s s i o n of models of speech production i n Section 1.6. Researchers, such as Abbs (1973), 14 Hardy (1970) and Bowman (1971; c f . Smith's 1973 review of Bowman), have c o n t r i b u t e d most of the informat i o n discussed i n t h i s s e c t i o n . The aud i t o r y system as a feedback mechan-ism f o r speech i s a l s o discussed here, and the main body of t h i s s e c t i o n ' s p r e s e n t a t i o n comprises an i n t r o d u c t i o n to the gamma, or s p i n d l e , motor system and i t s import f o r feedback. 1.31 The Gamma/Spindle Motor System "The gamma-loop can be considered as the gamma e f f e r e n t f i b e r s , the s p i n d l e f i b e r c o n t r o l l e d by the e f f e r e n t f i b e r s , and the synapt i c connections made by the s p i n d l e a f f e r e n t w i t h alpha motoneurons ( a f t e r Smith, 1969)." (Abbs, 1973, p. 176) A muscle s p i n d l e i s a small c y l i n d r i c a l body to the main body of the muscle. I t contains i n t r a f u s a l and e x t r a f u s a l muscle f i b r e s . The motor i n n e r v a t i o n of the body of a muscle i s c a r r i e d out by alpha motoneurons i n the co r t e x , while i n -n e r v a t i o n of muscle s p i n d l e s i s c a r r i e d out by gamma moto-neurons. Sensory neurons i n the muscle s p i n d l e form a mono-syn a p t i c r e f l e x arc w i t h alpha motoneurons. I t i s i n s t r u c t i v e at t h i s p o i n t to c i t e d i r e c t l y from Smith's (1973) summary of the major points i n Bowman's (1971) work concerning the gamma motor system: "Most muscles, i n c l u d i n g many but not a l l of the muscles innervated by the c r a n i a l nerves, contain small f u s i f o r m receptors known as muscle s p i n d l e s . The s p i n d l e s are l o c a t e d mechanically i n p a r a l l e l w ith the muscle f i b e r s , which are c a l l e d e x t r a f u s a l 15 f i b e r s i n t h i s context. The spi n d l e sends i n -formation back to the c e n t r a l nervous system (CNS) over two types of f i b e r s : large diameter group l a f i b e r s , and somewhat smaller group I I f i b e r s . The a f f e r e n t i n f o r m a t i o n , which can a r i s e f o r example when the sp i n d l e i s s t r e t c h e d , i n d i c a t e s i n a r e l a t i v e sense both the length of the muscle (group l a and group I I f i b e r s ) and the rat e of change of muscle length (group l a f i b e r s o n l y ) . E x t r a f u s a l muscle f i b e r s r e c e i v e t h e i r motor i n n e r v a t i o n from large d i a -meter alpha f i b e r s , which are axons of alpha motoneurones i n the s p i n a l cord or brainstem motor nucleus. Spindles also contain muscle ( c o n t r a c t i l e ) f i b e r s , c a l l e d i n t r a f u s a l f i b e r s . They r e c e i v e t h e i r i n n e r v a t i o n from gamma f i b e r s (axons of gamma motoneurones). The gamma i n -n e r v a t i o n of the s p i n d l e i s a complex i s s u e , and i s not at present completely understood. The alpha and gamma motor systems i n mammals are anatomically and f u n c t i o n a l l y d i s t i n c t . The s p i n d l e . w i t h i t s a s s o c i a t e d a f f e r e n t and motor nerve s u p p l i e s i s considered to be one of the most important mechanisms r e g u l a t i n g the s t a b i l i t y and accuracy of muscle c o n t r a c t i o n . An accurate understanding of the f u n c t i o n i n g of t h i s system i s thus obviously important f o r d e t a i l e d neuro-muscular s t u d i e s of the speech production appara-t u s . " (Smith, 1973, p. 172) Of s i m i l a r import i s the work of Merton (1953; c i t e d by Abbs, 1973), who claims that the s p i n d l e motor system can pro-duce output p r o p o r t i o n a l to the length e r r o r between e x t r a -f u s a l and i n t r a f u s a l f i b r e systems. The e r r o r s i g n a l i s 16 t r a n s m i t t e d to the motoneurons of the e x t r a f u s a l f i b r e s as "negative feedback". The muscle length i s then i n d i r e c t l y c o n t r o l l e d by c o n t r a c t i o n of the s p i n d l e f i b r e s . Abbs (1973) goes on to note that a common misunderstanding of the s p i n d l e motor system f o r feedback i n speech i s that many i n v e s t i g a t o r s f e e l i t to be a p e r i p h e r a l neuromotor network. In support of t h i s c o n t e n t i o n , Abbs c i t e s Mortimer and Akert's (1961) f i n d -ings from research w i t h primates which confirms that "gamma motoneurons have d i s c r e t e areas of c o r t i c a l r e p r e s e n t a t i o n very s i m i l a r to those of alpha motoneurons and i n some cases the two types are e x c i t e d by the same c o r t i c a l region. Such r e p r e s e n t a t i o n suggests-, not a d i f f u s e f a c i l i t o r y a c t i o n from the s p i n d l e motor system, but a d e t a i l e d c o r t i c a l l y c o n t r o l l e d f u n c t i o n " (Abbs, 1973, p. 178). I t i s perhaps the r o l e of the cerebellum, which receives a f f e r e n t muscle impulses from the brainstem and c o r t e x , to coordinate motor and sensory a c t i v i t y ( c f . Ruch et a l . , 1967; c i t e d i n Abbs, 1973, p. 178). From p a t h o l o g i c a l conditions of the cerebellum i t has been noted to c o n t r o l p r e c i s i o n of r a t e , range, force and d i r e c t i o n of voluntary motion. In sura: "the s p i n d l e motor system i s - n o t simply a p e r i p h e r a l c o n t r o l mechanism that serves only to modulate more c e n t r a l a c t i v i t i e s . The [ a v a i l a b l e ] data ... would suggest that t h i s system has the f u n c t i o n a l repre-s e n t a t i o n to i n t e r a c t w i t h alpha motoneuron systems i n the c o r t i c a l and s u b c o r t i c a l generation of speech movements. Indeed, Granit , as e a r l y as 1955 , sug-gested existence of separate but i n t e r a c t i n g c o n t r o l 17 of gamma and alpha motoneurons at higher neuromotor ce n t e r s . " (Abbs, 1973, p. 178) Abbs mentioned the h e s i t a t i o n of many speech-production researchers to assign the r o l e of " d e t e c t i o n and c o r r e c t i o n " of speech e r r o r s to the gamma-loop because of the r e l a t i v e l y long delay time i n v o l v e d . The c o n t r i b u t i o n of primary s p i n d l e a f f e r e n t f i b r e s and the s p e c i f i c r o l e of the spindle.motor system i n movement c o n t r o l are considered. As mentioned above, group l a a f f e r e n t spindle s r e l a y information on rate of length change of the muscle, and thus length can be a n t i c i p a t e d and problems of overshoot or o s c i l l a t i o n avoided. Consideration of the r o l e of these group l a f i b r e s , e s p e c i a l l y that of feed-back delay, leads to an awareness of a ser v o f u n c t i o n f o r the gamma e f f e r e n t system ( c f . Abbs, 1973, p. 179). In t h i s re-gard Fairbanks (1954) p o s t u l a t e d a model f o r speech i n terms of a servosystem (see a l s o Section 1.6 below); such a model compares output to input and adjusts input a c c o r d i n g l y i n t h i s closed-loop system. Von Eu l e r (1966; c i t e d i n Abbs, 1973, p. 179) supports the not i o n of a servosystem w i t h work on muscle s p i n d l e s i n the i n t e r c o s t a l muscles of the chest, i n which f i r i n g r ate of alpha motoneurons increases w i t h increased rate of r e s p i r a t i o n . Thus, "A continuous e r r o r s i g n a l that modulates c e n t r a l l y generated alpha a c t i v i t y could provide a b a s i s f o r the continuous c o r r e c t i o n of intended muscle length as set by the independently a c t i v a t e d gamma motor f i b e r s . " (Abbs, 1973, p. 179) 18 In s p i t e of i t s p r e d i c t i v e powers, a simple servo model cannot e x p l a i n continuous c o n t r o l i n production of short r a p i d muscle movements, such as those necessary f o r speech. The delay "loop" i n humans i s around 20-80 ms (c f . Campbell, 1968; A l s t o n et a l . , 1967; Sears § Newsome Davis, 1968: a l l c i t e d i n Abbs, 1973, p. 179), and some speech movements are completed i n l e s s time, such as those i n v o l v i n g tongue t i p , which are often i n i t i a t e d and completed i n l e s s than 50 ms. Stark (1968; c i t e d i n Abbs, 1973, p. 180) has suggested that the s p i n d l e (gamma) system i s used when "continuous" con-t r o l i s r e q u i r e d , and the alpha system may operate when high speed or b a l l i s t i c - t y p e movement i s requi r e d . Research by many i n v e s t i g a t o r s ( f o r d e t a i l s , see Abbs, 1973, pp. 180-181) l e d Abbs to summarize the p o s s i b l e r o l e of the sp i n d l e motor system as a " v a r i a b l e " servo model, i n the f o l l o w i n g terms: "the s p i n d l e motor system operates (1) when the muscle i s d i s t u r b e d i n an i s o m e t r i c s t a t e [ i . e . , when the ends of the muscle are f i x e d i n place and increase i n tension occurs without appreciable i n -creases i n length] or during c o n t r o l l e d i s o t o n i c c o n t r a c t i o n [ i . e . , when the t o t a l muscle i s of equal tension] by an unexpected force (that i s , the spi n d l e system attempts to maintain length or a c e r t a i n rate of change i n length) , (2) to develop speed i n the i n i t i a t i o n of c o n t r a c t i o n , and (3) to provide anta-g o n i s t i c f a c i l i t a t i o n to damp movement and prevent overshoot." (Abbs, 1973, p. 181) 19 1-32 A u d i t o r y Feedback The f i r s t suggestion that a u d i t o r y feedback may be i n -volved i n speech monitoring and p r o d u c t i o n , i n terms of the e f f e c t s of delayed auditory feedback on speech, was suggested by Lee (1950, 1951) and Black (1951). Other speech production models employing a u d i t o r y feedback were proposed by Fairbanks (1954) and Chase (1958). Their models w i l l be discussed be-low i n S e c t i o n 1.6. Van Riper (1971) speculates about the r o l e of auditory feedback f o r speech production. Acknowledging the contro-versy between continuous and i n t e r m i t t e n t monitoring of speech, Van Riper views speech as operating l i k e a servosystem under ordinary c o n d i t i o n s and claims t h a t : "Information about the speech output i s returned to the c e n t r a l i n t e g r a t i n g mechanism through s i x auditory channels, v i a the r i g h t and l e f t feedback routes from (1) airborne s i d e - t o n e , (2) bone-conducted sid e - t o n e , and (3) t i s s u e connected side-tone. Other feedback s i g n a l s come from the k i n e s t h e t i c - t a c t i l e -p r o p r i o c e p t i v e sensors on both sides of the body. Stromsta (1962) showed that auditory feedback s i g n a l s i n these d i f f e r e n t channels a r r i v e at markedly d i f f e r -ent times and t h a t the temporal information-processing of speech output by the b r a i n i s very complex. Some c e n t r a l mechanisms f o r i n t e g r a t i n g a l l these feedback s i g n a l s must be present, although t h e i r nature i s not yet known." (Van R i p e r , 1971, p. 383) Hardy (1970) discussed the importance of auditory moni-t o r i n g f o r maintenance of speech production. He considered 20 the c o n g e n i t a l l y deaf who do not u s u a l l y develop i n t e l l i g i b l e speech and the a d v e n t i t i o u s l y deaf ( i . e . , deafened a f t e r l e a r n i n g to speak) who show a slow d e t e r i o r a t i o n of the speech process ( c f . S a t a l o f f , 1966; c i t e d by Hardy, 1970, p. 50). Hardy supported the b e l i e f ( a f t e r Chase, 1958, and others) that the a u d i t o r y s i g n a l i s part of the t o t a l sensory exper-ience used i n generating speech w i t h complete inform a t i o n about speech musculature p a t t e r n i n g contained i n the speech " t a r g e t " . Lombard ( c i t e d i n Hardy, 1970, p. 51), who blocked audi-tory feedback by masking the s u b j e c t s ' speech w i t h high i n -t e n s i t y n o i s e , found that -- other than r a i s i n g the i n t e n s i t y l e v e l of the voice -- speakers showed l i t t l e d i s r u p t i o n of a r t i -c u l a t i o n . This l e d Hardy to r e j e c t Fairbanks's (1954) model of closed-loop feedback f o r speech, where the system would be dependent on a u d i t o r y feedback at a l l times. Perhaps the speaker-, i n r a i s i n g h i s v o i c e , i s a d j u s t i n g h i s auditory feed-back l e v e l so t h a t i t i s audible under such c o n d i t i o n s , at l e a s t v i a bone conduction. The a d v e n t i t i o u s l y deafened speaker does not l o s e speech immediately, i n s p i t e of the auditory feedback having been l o s t . Hardy's conclusion may shed some l i g h t on these c o n s i d e r a t i o n s : " i t must be concluded that i n t r a o r a l sensations can provide cues f o r p o s i t i o n i n g of the speech muscula-ture once the appropriate p a t t e r n i n g has been learned, and they can continue to do so i n the absence of auditory feedback." (Hardy, 1970, p. 51) 21 1.33 Summary To synthesize i n f o r m a t i o n from Section 1.2 on neuro-physiology and from t h i s s e c t i o n , a s p e c u l a t i v e comment from Konigsmark (1970) , based on h i s knowledge of neural s t r u c -tures and t h e i r connections, i s appropriate at t h i s time: "The i n t e g r a t i v e a c t i v i t y r e s u l t i n g i n speech probably begins i n the c e r e b r a l cortex w i t h a con-cept which can be v o c a l i z e d . Broca's area i n the cortex may then be i n f l u e n c e d to i n i t i a t e the speech process. P r o j e c t i o n s from t h i s c o r t i c a l region go to the motor cortex. From here a major p r o j e c t i o n courses to the motor n u c l e i i n v o l v e d i n speech, that i s , the hypoglossal nucleus, the nu-cleus ambiguus, the f a c i a l nucleus, and the motor nucleus of the Vth nerve. At the same time, f i b e r s from the c e r e b r a l cortex p r o j e c t to the basal gang-l i a , and to the c e r e b e l l a r cortex v i a the po n t i s [pons]. These p r o j e c t i o n s probably f u n c t i o n to smooth and to create the necessary motor tonus f o r v o c a l i z a t i o n . P r o j e c t i o n s from Broca's area to the r e s p i r a t o r y motor area may coordinate t h i s a c t i -v i t y w i t h speech. "Neurons i n the hypoglossal nucleus, nucleus ambiguus, f a c i a l nucleus, and motor nucleus of the Vth nerve are played upon by p r o j e c t i o n s from the p r e c e n t r a l gyrus and by the c e r e b e l l a r cortex v i a the red nucleus. A l s o , s h o r t e r connections i n t e r -connect these n u c l e i w i t h one another, p o s s i b l y a i d i n g i n t h e i r coordinated a c t i v i t y . Fibers from these n u c l e i act upon the musculature of the tongue, l a r y n x , mouth, and jaw to produce speech. "Sensory endings i n the mucosa and musculature of the tongue, l a r y n x , mouth, and jaw are a c t i v a t e d by touch, pressure, and p o s i t i o n . This information 22 i s fed i n t o the d o r s a l horns of the s p i n a l cord, the nucleus s o l i t a r i u s , and to the t r i g e m i n a l sensory complex. These s t r u c t u r e s are also i n -fluenced by the c e r e b r a l cortex and r e t i c u l a r formation, p o s s i b l y enhancing or dampening t h e i r a c t i v i t y , as the occasion demands. These sen-sory n u c l e i then p r o j e c t to the v e n t r a l p o s t e r i o r medial nucleus of the thalamus, and then to the p o s t c e n t r a l gyrus of the cortex. "Auditory feedback of what i s being s a i d pro-j e c t s to the transverse temporal g y r i . From these g y r i there are p r o j e c t i o n s to the motor co r t e x , a l l o w i n g a comparison of the r e s u l t s of speech and p o s s i b l y i n f l u e n c i n g the motor production of speech." (Konigsmark, 1970, p. 17) 1. 4 Speech E r r o r s As defined by Boomer and Laver (1968) , a " s l i p of the tongue ... i s an i n v o l u n t a r y d e v i a t i o n i n performance from the speaker's current p h o n o l o g i c a l , grammatical or l e x i c a l i n t e n t i o n " (Fromkin, 1971, p. 29). L i n g u i s t i c a l l y , speech e r r o r s have been stud i e d f o r sev-e r a l reasons: (1) to provide important clues f o r language change, to provide a source f o r studying h i s t o r i c a l l i n g u i s t i c change, as suggested by Hermann Paul (Sturtevant, 1917; MacKay, 1970); (2) to understand b e t t e r the speech production process v i a the mechanisms of speech (Boomer § Laver, 1968; Fromkin, 1968; Nooteboom, 1969); and (3) to draw a d i s t i n c t i o n between "competence" and "performance", and to demonstrate the r e a l i t y of p h o n o l o g i c a l u n i t s and r u l e s (Fromkin, 1968). 23 Speech e r r o r s can occur whenever speech i s used. Meringer (1908) recorded, along w i t h the speech e r r o r s , the speaker's b i r t h d a t e , h i s educ a t i o n a l background, s t a t e of h e a l t h , degree of t i r e d n e s s , rate of speech, and the time of day at which such e r r o r s occurred, only to f i n d that there was no c o r r e l a t i o n of any of these f a c t o r s with the err o r s ob-served. Several i n v e s t i g a t o r s have devoted time to c l a s s i f y i n g speech e r r o r s i n t o such cat e g o r i e s as misordering, omission or replacement of a u n i t (Boomer § Laver, 1968), or i n t o phon-emic yjs non-phonemic e r r o r s (Nooteboom, 1969). Fromkin (1971) considers e r r o r s , not f o r purposes of c l a s s i f i c a t i o n , but as evidence f o r un d e r l y i n g u n i t s i n speech, such as s y l l a b l e , phoneme and f e a t u r e . From Fromkin's (1968) research on speech e r r o r s , one can see that such e r r o r s obey r u l e s of the grammar and are not ran-domly generated. Her r e s u l t s show t h a t : (1) f e a t u r e s , seg-ments and s y l l a b l e s make up u n i t s of pro d u c t i o n ; (2) segments i n a s y l l a b l e are ordered, and t h i s order i s not v i o l a t e d i n the production of an e r r o r ; (3) morphemes or words of the same cla s s ( i . e . , r o o t s , a f f i x e s , etc.) are u s u a l l y interchanged with one another; (4) i n t o n a t i o n and primary word s t r e s s re-main i n the same p o s i t i o n , regardless of the e r r o r ; (5) mor-p h o l o g i c a l and phonetic or ph o n o l o g i c a l c o n s t r a i n t s are placed on a word at d i f f e r e n t times i n the generation of an utterance; (6) non-permissible phonetic sequences ( i . e . , those not char-a c t e r i s t i c of the language) do not occur; (7) semantic . 24 features may be d i s p l a c e d , r e s u l t i n g i n a semantic e r r o r ; and (8) words w i t h s i m i l a r i t y of p h o n o l o g i c a l form are l i k e l y candidates f o r s u b s t i t u t i o n as an e r r o r ( c f . Fromkin, 1971). As.Fromkin (19 71) and MacKay (1970) have noted, speech e r r o r s are more l i k e l y to occur between words that contain s i m i l a r phonetic elements. I t has also been noted that re-s u l t a n t e r r o r s of metathesis of two s e q u e n t i a l phonemes i n words (e . g. , /assk/ -> /aeks/) of ten seem to i n v o l v e the s i b i l a n t / s / : "In a number of p e r c e p t i o n t e s t s , the h i s s (such as occurs w i t h [s] i s o f t e n 'misplaced'; i . e . , i t i s d i f f i c u l t f o r subjects to judge where the noise occurs i n an utterance" (Fromkin, 1971, p. 39). On the b a s i s of such statements (and the data) one may suggest that words of s i m i l a r p honological make-up i n v o l v i n g the s i b i l a n t / s / are the most l i k e l y to create speech e r r o r s . Of the s t u d i e s c a r r i e d out on speech e r r o r s , those of Meringer and Mayer (1895) and Meringer (1908) are the most extensive i n terms of number of e r r o r s and p o s s i b l e e x t r a -l i n g u i s t i c c o r r e l a t i o n s . Boomer and Laver (1968) c o l l e c t e d over one hundred e r r o r s , and Fromkin (1971) reported over s i x hundred e r r o r s ; but only Boomer and Laver tape-recorded t h e i r e r r o r s . Fromkin c o l l e c t e d hers i n an anecdotal f a s h i o n , g e n e r a l l y w i t h the speaker r e p o r t i n g a f t e r the f a c t what he had s a i d and had meant to say. One major problem w i t h From-ki n ' s method of data c o l l e c t i o n f o r a d i s c u s s i o n at the mole-c u l a r l e v e l of speech production i s that s u b t l e phonetic 25 d i f f e r e n c e s , or d e v i a n c i e s , i n an e r r o r w i l l (at l e a s t some-times) be missed by the speaker and hence not reported, lead-ing to f a l s e claims about the phonological/phonetic form --and perhaps cause -- of a speech e r r o r . In summary, speech e r r o r s have been used i n various i n -v e s t i g a t i o n s to t e s t d i v e r s e hypotheses, but no e m p i r i c a l ex-amination of the e r r o r phenomena has i t s e l f been conducted. As a r e s u l t of such c o n s i d e r a t i o n s , i t was f e l t that an exam-i n a t i o n of durations of segments i n speech e r r o r s , s p e c i f i c a l l y i n w o r d - i n i t i a l consonant c l u s t e r s -- an aspect of the problem not i n v e s t i g a t e d by Meringer or others -- would prove informa-t i v e and help elaborate hypotheses concerning the genesis of such e r r o r s . 1.5 Timing of Speech Timing, f o r the purposes of t h i s study, i s defined as a s e q u e n t i a l o r d e r i n g of a r t i c u l a t o r y events, t h e o r e t i c a l l y based on neuromuscular impulses sent by the b r a i n to the a r t i -c u l a t o r s and programmed i n some h y p o t h e t i c a l u n i t , such as a phoneme, morpheme, or s y l l a b l e . I t i s not known from the l i t e r a t u r e whether timing of speech i s r e g u l a r ; i . e . , whether i t remains constant, f o r ex-ample, from s t r e s s to s t r e s s w i t h i n an utterance (cf. Ohala, 1970). Nor i s i t known when timing begins; i . e . , whether a timing programme i s set down when the f i r s t phoneme i s u t t e r e d 26 or as soon as the impulse i s i n i t i a t e d i n the b r a i n . Ohala (1973) gives r e s u l t s which support the claims that there i s no u n d e r l y i n g speech rhythm or time programme, as claimed by Kozhevnikov and C h i s t o v i c h (1965). A number of i n v e s t i g a t o r s have considered timing i n at-tempts to construct models of speech p r o d u c t i o n , together w i t h concomitant c o a r t i c u l a t o r y e f f e c t s (e.g., Haggard, 1973; Kent $ M o l l , 1975 ; L e h i s t e , 1971; Ohala, 1970 , 1973). These models can u s u a l l y be c l a s s i f i e d i n terms of systems which do or do not employ c o n s i d e r a t i o n s of feedback i n speech pro-duction (see also Section 1.6). Timing i s considered i n t h i s study only i n s o f a r as i t i s a p o s s i b l e determinant of the d u r a t i o n of segments which, i t i s t e n t a t i v e l y hypothesized here, i t c o n s t r a i n s . 1.6 Models of Speech Production 1.60 I n t r o d u c t i o n Research on speech e r r o r s , t i m i n g of speech and feedback has l e d to the f o r m u l a t i o n o f numerous t h e o r i e s and models of speech p r o d u c t i o n , s e v e r a l o f which were b r i e f l y mentioned e a r l i e r . In g e n e r a l , models of speech production can be c l a s s i f i e d as e i t h e r " c l o s e d - l o o p " or "open-loop": the former r e f e r s to a system which -- i t i s speculated -- employs feed-back to regu l a t e and adjust speech, the l a t t e r to a system which does not. The closed- vs open-loop d i s t i n c t i o n has been 27 v a r i o u s l y designated as a " c h a i n " vs_ "comb" model ( B e r n s t e i n , 1967), " s e q u e n t i a l chain" vs_ " p l a n " model (Lenneberg, 1967), "hypothesis 1" vs_ "hypothesis 2", using e f f e r e n t and a f f e r e n t impulses (Kozhevnikov § C h i s t o v i c h , 1965) , and " p e r i p h e r a l feedback or c h a i n i n g " vs_ "preprogramming" model (Kent § M o l l , 1975). Our d i s c u s s i o n begins w i t h c o n s i d e r a t i o n of the theo-r i e s which support an open-loop hypothesis. 1.61 Open-Loop Models As noted above, an open-loop system s p e c i f i e s that com-mands are issued to the a r t i c u l a t o r s at r e g u l a r i n t e r v a l s to produce speech, but that feedback i s not employed to regulate i t s p roduction. One of the more in n o v a t i v e studies o r i g i n a t e d w i t h Kozhevnikov and C h i s t o v i c h (1965) , who t e s t e d the r a t i o n -ale f o r using e i t h e r of the two models to account f o r the se-q u e n t i a l generation of s y l l a b l e s . D e f i n i n g a "syntagma" as a sentence or phrase connected by meaning and a r t i c u l a t i o n and pronounced on a s i n g l e output, they v a r i e d the rate of speech and s t r e s s of the syntagma and found the s y l l a b l e to be the u n i t which remained r e l a t i v e l y constant under such v a r i a -t i o n . T h e i r hypothesis was that i f an open loop i s i n opera-t i o n , the t o t a l variance over the time i n t e r v a l w i l l be l e s s than the sum of the variances of the component i n t e r v a l s , and t h e i r data support t h i s c l a i m . MacNeilage (1970) i n i t i a l l y supports an open-loop system, cl a i m i n g that command patterns issued to the muscles would not wait f o r inf o r m a t i o n from the a r t i c u l a t o r reaching a given 28 t a r g e t i n order to c o n t r o l the f o l l o w i n g movement appropri-a t e l y . However, he points out that an open-loop system would r e q u i r e storage o f a vast amount of i n f o r m a t i o n on phoneme-to-phoneme t r a n s i t i o n s ; and s i n c e t h i s i s not p r a c t i c a l , he sug-gests that closed-loop c o n t r o l i s more probable, based on n e u r o p h y s i o l o g i c a l research on the gamma motor system (cf. S e c t i o n 1.3). Fromkin's (1971) model of speech production shows a pos-s i b l e o r d e r i n g of events i n generation of an utterance and accounts f o r p r o d u c t i o n of e r r o r s , as w e l l as c o r r e c t u t t e r -ances, but i t shows no r e l a t i o n s h i p of these w i t h any feed-back mechanism. Since there i s no mention of feedback, espe-c i a l l y concerning e r r o r utterances i n which the e r r o r i s "caught" and then c o r r e c t e d , i t can be assumed that Fromkin's model i s more c l o s e l y a s s o c i a t e d w i t h an open-loop than, w i t h a closed-loop system. 1.62 Closed-Loop Models The e a r l i e s t notions of a closed-loop model, f o l l o w i n g c y b e r n e t i c theory ( i . e . , a system which employs feedback by which to modify subsequent productions w i t h i n a given u t t e r -ance) are provided by Lee (1950) and Fairbanks (1954). Lee proposed a system of loops: a r t i c u l a t i o n loops, monitoring phonemes v i a t a c t i l e and k i n e s t h e t i c means, and voice loops, monitoring s y l l a b l e s v i a a u d i t o r y feedback; both of these operate on v o l i t i o n and r e f l e x systems. Fairbanks i n t e r -preted the speech system as a closed-loop servosystem, i n 29 which the output i s compared w i t h input v i a bone-conducted and air-conducted auditory feedback and which manipulates the p r o d u c t i o n mechanism so t h a t output w i l l have the same f u n c t i o n a l form as input. Chase's (1958) model of speech production l i k e w i s e i n -corporates a servosystem and auditory feedback and evinces the same flaw as Fairbanks's model, namely that monitoring speech s o l e l y by means of a u d i t o r y feedback would mean that speaking would be impossible (or at l e a s t i n o r d i n a t e l y d i f -f i c u l t ) i n an extremely nois y environment, since feedback would be e f f e c t i v e l y masked. As many who work i n i n d u s t r i a l environments where there are e x c e p t i o n a l l y high noise l e v e l s w i l l a t t e s t , t h i s i s not the case. Kent and M o l l (1975) p o s i t a feedback model which assumes that timing of movements from higher centres depends on ef-ferent and a f f e r e n t s i g n a l s r e c e i v e d from a previous a r t i -c u l a t o r y movement, i n order to chain together speech segments. Their "preprogramming", or open-loop> model assumes that inher-ent timing c o n t r o l r e s u l t s i n the timing of an a r t i c u l a t o r y movement's being a f f e c t e d by another a r t i c u l a t o r y movement which occurs e i t h e r before or a f t e r i t . While they i n i t i a l l y i n t e r p r e t the data from t h e i r i n v e s t i g a t i o n of w o r d - i n i t i a l / s p r - / and / s p l - / c l u s t e r s i n terms of a closed-loop system, Kent and M o l l note that responses from t h i s feedback loop f a i l to reach consciousness and t h a t a r t i c u l a t i o n must there-fore depend on "unconscious feedback-mediated responses" (1975, p. 319); i t i s not c l e a r from t h e i r argument how feedback 30 might ever be a conscious response. They reconsider t h e i r data i n terms of an open-loop model, which they cl a i m i s the only way to account f o r : (a) v a r i a b l e duration of / s / before /p/ i n /sp-/ c l u s t e r s , and (b) the ordering of / s / - r e l e a s e upon the gesture f o r / p / - c l o s u r e , l i k e w i s e i n /sp-/ c l u s t e r s . However, they do not say how an open-loop system would accom-p l i s h t h i s : No r e s o l u t i o n i s forthcoming, except f o r the i n c i -d ental non-comment t h a t , whatever model one adopts, i t w i l l be "capable of f i n e and accurate c o n t r o l " (Kent 5 M o l l , 1975, p. 321). 1.63 Summary and D i s c u s s i o n Ohala (1970) t r i e s to c l e a r up some of the misconceptions surrounding the r e s u l t s and i n t e r p r e t a t i o n of Kozhevnikov and C h i s t o v i c h 1 s (1965) experiment, commenting that the methodo-logy employed cannot r e v e a l adequate information as to the presence or absence of feedback i n the timing of speech. He proposes a r e i n t e r p r e t a t i o n of t h e i r closed- and open-loop hypotheses: The f i r s t i s a system which i s "'Timing Dominant' ... i . e . , a system which maintains a t i g h t time schedule per-haps at the expense of p r e c i s e and thorough accomplishment of the gestures"; the second i s one which i s " ' A r t i c u l a t i o n Dominant' ... i . e . , a system which maintains p r e c i s e and thor-ough performance of the gestures no matter how much time i t takes" (Ohala, 1970, p. 143). He adds that both of these systems could e i t h e r employ or not employ feedback i n deter-mining future a r t i c u l a t o r y events. 31 The major c r i t i c i s m l e v e l e d by Ohala (1970) against the methodology used by Kozhevnikov and C h i s t o v i c h (1965), i n which subjects were re q u i r e d to repeat the same utterance hundreds of times at d i f f e r e n t rates of speech, i s that per-haps the subjects adopted a f i x e d rhythm and t h i s could a f f e c t the u n d e r l y i n g time schedule and generation of speech segments. Under such c o n d i t i o n s an open-loop model, where timing commands are sent at f i x e d i n t e r v a l s , i s more l i k e l y to be adopted as an ex p l a n a t i o n of t h e i r f i n d i n g s . In terms of t h e i r data a n a l y s i s , Ohala has c r i t i c i z e d Kozhevnikov and C h i s t o v i c h f o r t h e i r r e l i a n c e on p o s i t i v e and negative covariances between phones i n the determination of which model should be chosen; by covariance, here i s meant that i f an e r r o r i s made i n the duration of a phone, e i t h e r p o s i t i v e l y or n e g a t i v e l y , i t i s compensated f o r i n the f o l l o w -ing phone, which f i n i s h e s at the o r i g i n a l l y planned time, by being e i t h e r shortened or lengthened, r e s p e c t i v e l y . Ohala (1973) claims that v a r i a t i o n s i n rate of speech of the t e s t items, i f c o n s i s t e n t l y y i e l d i n g p o s i t i v e covariances, would tend to support an open-loop model, which i s indeed the one adopted by Kozhevnikov and C h i s t o v i c h ; i f c o n s i s t e n t l y nega-t i v e covariances were obtained, on the other hand', a closed-loop model would suggest i t s e l f . Ohala (1970) also questions the appropriateness of the closed-loop system f o r speech. He presents s e v e r a l arguments i n favour of such a system and then proceeds to disprove a l l of them. He does, however, confirm the p o s s i b i l i t y of the 32 use of short-term feedback to make quick adjustments i n speech (by recourse to the r e s u l t s of h i s experiment on maxi-mum j aw opening and v e l o c i t y i n the production of i s o l a t e d words; c f . Ohala, 1970, pp. 122-141). A model of speech production based on n e u r o p h y s i o l o g i c a l mechanisms which incorporates both open- and closed-loop sys-tems has been proposed by Abbs (1973). He views the com-p l e t e system as a " v a r i a b l e " servosystem ( i . e . , one which can employ feedback depending on the system's requirements). He claims that the gamma motor system ( c f . Section 1.31) i n -volved i n feedback: (1) maintains the length or rate of change of length of a muscle, (2) i s a c t i v e i n i n i t i a t i o n of c o n t r a c t i o n , and (3) provides a n t a g o n i s t i c actions to damp movement and prevent overshoot. A speech per c e p t i o n / p r o d u c t i o n model which i s an a c t i v e a n a l y s i s process a p p l i e d to the speech s i g n a l i s the a n a l y s i s -by-synthesis model proposed by B e l l et a l . (1961). The main part of t h i s system i s a generator capable of s y n t h e s i z i n g a l l s i g n a l s to be analysed. These s i g n a l s are compared w i t h s i g n a l s to be analysed and an e r r o r measure computed. When a s i g n a l i s sy n t h e s i z e d that causes the e r r o r to reach the smallest value, t h i s s i g n a l i s stored. Components of the sys-tem are the f i l t e r s e t , spectrum generator, comparator, and st r a t e g y component. The designers claim that t h i s system re-presents l i n g u i s t i c phenomena at various l e v e l s such as a c o u s t i c , p h o n o l o g i c a l , morphological and s y n t a c t i c . This 33 system i s mentioned here because i t considers p r o d u c t i o n , p e r c e p t i o n , and feedback of speech a l l i n one e f f i c i e n t model. On the assumption that feedback may or may not be i n -volved i n the production of speech (depending on various con-d i t i o n s , as yet unknown), arguments f o r one type of system or another are perhaps premature, given the lack of a proper e m p i r i c a l foundation. Much more research needs to be c a r r i e d out i n search of an answer to the problem, and a t t e n t i o n should now be devoted to d e v i s i n g experiments which can ade-quately t e s t f o r t h i s i n t e r m i t t e n t feedback and to determining the r o l e i t plays i n speech production. CHAPTER 2 STATEMENT OF THE PROBLEM There are few st u d i e s which examine models of speech production i n conjunction w i t h t h e i r p o s s i b l e neurophysio-l o g i c a l bases, and even fewer i n number are those i n v e s t i -gations which incorporate c o n s i d e r a t i o n s of feedback mechan-isms i n t o such models. I f we consider production of a speech e r r o r as a momen-tar y breakdown, followed by a r e c a l i b r a t i o n of the system enabling the c o r r e c t i o n of an e r r o r , and i f we use such e r r o r s to hypothesize about c e r t a i n aspects of speech pro- . duction ( i n c l u d i n g feedback and n e u r o p h y s i o l o g i c a l mechan-isms) , then perhaps i t might be p o s s i b l e to provide f u r t h e r i n s i g h t i n t o the process of speech production. Timing of speech i s considered i n r e l a t i o n to speech e r r o r s f o r the purpose of determining d i f f e r e n c e s i n d u r a t i o n of segments i n r e p e t i t i o n e r r o r s (of the form /s*C*/ ... /sC/), between /s*/ and /C*/ i n an e r r o r production and / s / and /C/ i n a c o r r e c t l y produced r e p e t i t i o n immediately f o l l o w i n g the e r r o r . Because the e r r o r i s corrected immediately f o l l o w i n g i t s commission (although there may be a s l i g h t h e s i t a t i o n , to be described i n Chapter 4 ) , rat e of speech i s considered to remain constant and the r e f o r e not to a f f e c t s i g n i f i c a n t l y the length of consonants. Based on the foregoing considera-t i o n s , comparisons are made between /s*/ and / s / and between /C*/ and /C/. 34 35 The i n t e n t of the present study was to examine systemati-c a l l y s p e c i f i c aspects of speech er r o r s through studying re-l a t i o n s h i p s of w o r d - i n i t i a l f r i c a t i v e plus stop consonant c l u s t e r s , by: (1) determining an e f f i c i e n t procedure f o r generating speech e r r o r s and f o r t h e i r subsequent a n a l y s i s ; (2) o b t a i n i n g a r e p r e s e n t a t i v e sampling of speech errors' i n w o r d - i n i t i a l f r i c a t i v e plus stop con-son an t c 1 us t e rs ; (3) p r o v i d i n g a sample of speech at normal conversa-t i o n a l rate c o n t a i n i n g no speech e r r o r s ; (4) c l a s s i f y i n g the types of speech e r r o r s found by c o n s u l t i n g a v a i l a b l e d e s c r i p t i v e accounts from previous research; (5) e v a l u a t i n g the duration and timing r e l a t i o n s h i p s of c e r t a i n of the w o r d - i n i t i a l c l u s t e r s under i n v e s t i g a t i o n ; (6) c o n s i d e r i n g the experimental r e s u l t s i n terms of various models of speech production, feedback mechanisms, and speech perception models. CHAPTER 3 METHOD 3.1 P i l o t Studies P r i o r to the main study, two p i l o t s t u d i e s were conducted i n order to a s c e r t a i n a reasonably optimal approach'to the c o l l e c t i o n and a n a l y s i s of speech e r r o r s . 3.11 P i l o t Study I Subj e c t s . Subjects f o r the f i r s t p i l o t study were four female u n i v e r s i t y s tudents, ranging i n age from 21 to 25 years. A l l were n a t i v e speakers of E n g l i s h w i t h no demonstrable hearing or speech problems. Stimulus M a t e r i a l s . Three types of stimulus m a t e r i a l s were used i n d i f f e r e n t combinations to determine which woi*ld pro-duce the most f a t i g u e of speech musculature and thus give the greatest number of speech e r r o r s : (1) The f i r s t paragraph of "The Rainbow Passage" ( c f . F a i r -banks, 1960, p. 127; see Appendix A). (2) Six occurrences of each of the c l u s t e r s /sp-/, / s t - / and /sk-/ i n i n i t i a l p o s i t i o n i n words embedded i n three separate paragraphs (see Appendix B). (3) Twelve occurrences of each of the w o r d - i n i t i a l c l u s t e r s s p e c i f i e d i n (2) above i n words embedded i n three sets of sentences, each r e f e r r e d to here as a "tongue-twister" (see Appendix C). 36 37 Procedure. Subject 1 was i n s t r u c t e d to read the "Rainbow Passage" at normal speed, then to read the three paragraphs f i v e times each c o n s e c u t i v e l y as f a s t as p o s s i b l e . Subject 2 was i n s t r u c t e d to read the "Rainbow Passage" at normal speed, and then to read the three "tongue-twisters" f i v e times each as f a s t as she could. Subject 3 was t o l d to read each of the paragraphs once at normal speed and then f i v e times each as f a s t as she could. F i n a l l y , Subject 4 was re-qu i r e d to read each of the "tongue-twisters" once at normal speed and then f i v e times each as q u i c k l y as p o s s i b l e . Resu l t s. Subject 1 produced three e r r o r s ; Subject 2, t h i r -teen e r r o r s ; Subject 3, nine e r r o r s ; and Subject 4, f i f t e e n e r r o r s . Thus, more err o r s were produced by those subjects who had read the "tongue-twisters" than those who had not. In the l i g h t o f these r e s u l t s , i t was deemed necessary to conduct a second p i l o t study using only "tongue-twisters" as stimulus m a t e r i a l s , i n order to r e f i n e the experimental procedures. 3.12 P i l o t Study I I Subj ect s . For t h i s study subjects were two female u n i v e r s i t y students, both 24 years of age, nati v e speakers of E n g l i s h with no demonstrable hearing or speech d i f f i c u l t i e s . Stimulus M a t e r i a l s . The stimulus m a t e r i a l s c o n s i s t e d of the three " t o n g u e - t w i s t e r s " described i n Section 2.11 above and given i n Appendix C. Each sentence contained twelve 38 occurrences of one of the w o r d - i n i t i a l c l u s t e r s /sp-/, / s t - / or /sk-/ . Procedure. Each subject was re q u i r e d to read each sentence f i f t e e n times as f a s t as p o s s i b l e . Only the l a s t f i v e r e p e t i -t i o n s were examined f o r speech e r r o r s , the f i r s t ten produc-t i o n s being considered as the " f a t i g u i n g " p o r t i o n of the experiment. Result s. The number of e r r o r s produced by subjects reading " t o n g u e - t w i s t e r s " ( i n both p i l o t s t u d i e s ) was greater by a f a c t o r of two to one than those obtained through the use of any other passage. As such, they were considered to produce b e t t e r " f a t i g u i n g " e f f e c t s and were therefore chosen as stim-ulus m a t e r i a l s f o r the experiment proper. 3.13 P i s cus s ion The types of e r r o r s produced by the subjects i n the p i l o t s t u d i e s could be c l a s s i f i e d as f o l l o w s : (a) s y l l a b l e r e p e t i -t i o n , (b) phonetic s u b s t i t u t i o n , (c) c l u s t e r r e p e t i t i o n , (d) phoneme r e p e t i t i o n , and (e) phoneme pr o l o n g a t i o n . These e r r o r s w i l l be discussed more f u l l y i n Chapter 4. I t was f e l t that reading e r r o r s might have c o n t r i b u t e d to the percentage of the e r r o r s produced, but there appeared to be no o b j e c t i v e way to e x t r a c t speech e r r o r s under such c o n d i t i o n s . In order to avoid the p o s s i b i l i t y of reading e r r o r s , given that the purpose of the i n v e s t i g a t i o n was to examine spontaneous productions, i t was decided to have the subjects i n the main 39 study memorize the "tongue-twisters" r a t h e r than read them. Such a method has been described and employed by Kozhevnikov and C h i s t o v i c h (1965). 3.2 Main Study Subjects. S e l e c t e d f o r i n c l u s i o n i n the study were twelve u n i v e r s i t y students, ranging from 22 to 33 years of age, a l l of whom were n a t i v e speakers of E n g l i s h . The hearing of a l l subjects was judged to be w i t h i n normal l i m i t s , and none had deviant a r t i c u l a t i o n p a tterns or anomalies of f a c i a l musula-tu r e . Equipment. The equipment employed i n t h i s study i n c l u d e d : (a) For r e c o r d i n g : a two-channel power supply ( B r l i e l 6j K j a e r , Type 2803), a s s o c i a t e d w i t h a B r l i e l § Kjaer one-inch micro-phone, Model 4145; and a S c u l l y Model 280-2 tape recorder/ reproducer. Recordings were made on Ampex 611 audiotape at 7.5 i p s . (b) For a n a l y s i s : a Kay Sona-Graph, Model 7029A, a Siemens O s c i l l o m i n k , and an Ampex Micro 50 cassette recorder/reproducer, u t i l i z i n g s t u d i o - q u a l i t y magnetic tape. Stimulus M a t e r i a l s and Procedure. The stimulus m a t e r i a l s em-ployed i n c l u d e d the three "tongue-twisters" described above and given i n Appendix C. Each subject was i n s t r u c t e d to memorize one sentence at a time. When the subject i n d i c a t e d that she knew the passage w e l l enough to r e c i t e i t aloud without e r r o r or prompting, 40 taping began, during which time a 12-inch mouth-to-microphone distance was maintained. I n s t r u c t i o n s to a l l subjects were as f o l l o w s : "Repeat the sentence once at normal c o n v e r s a t i o n a l r a t e i n a normal c o n v e r s a t i o n a l v o i c e , and then f i f t e e n times as f a s t as you can." The s u b j e c t s followed t h i s procedure f o r a l l three "tongue-t w i s t e r s " , whereby the order of p r e s e n t a t i o n of the three sentences was v a r i e d randomly f o r a l l s u b j e c t s . 3.3 A n a l y s i s of Data Tapes and T r a n s c r i p t i o n . Tapes were t r a n s c r i b e d f o r a l l twelve subjects by the experimenter using a modified v e r s i o n of the I n t e r n a t i o n a l Phonetic Alphabet (IPA). This a n a l y s i s included the f i r s t p r o d u c t i o n at normal speed and the f i f t e e n t e s t re-p e t i t i o n s , i n c l u d i n g any e r r o r s , together w i t h t h e i r immediate phonetic contexts. Tapes could be played back e i t h e r at nor-mal speed or at half-normal speed, i n order to f a c i l i t a t e t r a n s c r i p t i o n . E r r o r s were then coded (see Appendix D f o r examples of t r a n s c r i p t i o n and coding); and where necessary, spectrograms were produced to determine more c l e a r l y phonetic s u b s t i t u t i o n s , epenthetic phones, and r e v e r s a l s w i t h i n c l u s t e r s . E d i t i n g . Because of the extent of the data a v a i l a b l e , only data from the s i x subjects who made the most e r r o r s were e d i t e d and analyzed. Phonetic t r a n s c r i p t i o n s were v a r i f i e d by having a t r a i n e d p h o n e t i c i a n t r a n s c r i b e e r r o r s from these s i x s u b j e c t s . E d i t i n g was c a r r i e d out v i a the S c u l l y 280-2 tape recorder/ reproducer i n conjunction w i t h the Ampex Micro 50 cassette 41 recorder, the experimenter i s o l a t i n g the normal-conversational-rate r e p e t i t i o n of the "t o n g u e - t w i s t e r s " , a l l e r r o r s , and t h e i r immediate contexts.. Oscillograms. E d i t e d data were d i s p l a y e d v i s u a l l y on o s c i l l o -grams, using the Siemens O s c i l l o m i n k . This instrument d i s p l a y s a speech s i g n a l t r a c e , duplex o s c i l l o g r a m t r a c e , and a trace of the l o g of average speech power. Within the experimental arrangement i s a Revox Model A77 tape recorder/reproducer, duplex o s c i l l o g r a p h , FrszSkj aer-Jensen Trans-Pitchmeter, and an i n t e n s i t y or speech power c i r c u i t . Oscillograms were produced at ten cm/s. I n s e r t Figure 1 about here. Segmentation. Because of the r a p i d i t y of the s u b j e c t s ' speech, segmentation of the o s c i l l o g r a m s proved somewhat d i f f i -c u l t . Gross segmentation was c a r r i e d out f i r s t . This was accomplished by marking o f f 10-cm s e c t i o n s on the o s c i l l o g r a m . The utterance on the tape was then timed w i t h a stopwatch to a five-second mark, and at t h i s p o i n t the o s c i l l o g r a m was matched by counting working backwards from every f i v e to three to one second marks. When a subject's utterances were segmented at the gross l e v e l , a f i n e r segmentation procedure was conducted. One o b j e c t i v e , of the f i n e r segmentation was to e s t a b l i s h the time i n m i l l i s e c o n d s of the /s/-segments plus stop consonant TAPE RECORDER DUPLEX OSCILLOGRAPH TRANS-PITCHMETER SPEECH POWER METER OSCILLOMINK FIGURE 1. Block diagram of in s t r u m e n t a t i o n used to produce o s c i l l o g r a m s . 42 43 i n : (a) the e r r o r segment, and (b) the c o r r e c t productions of the utt e r a n c e . S e l e c t i o n of / s / + /C/ (stop) avoided many of the d i f f i c u l t i e s a s s o c i a t e d w i t h segmentation of g l i d e s , resonants, vowels and phonemes w i t h i n words which were e x c e s s i v e l y shortened, d i s t o r t e d or omitted because of the r a p i d i t y of the s u b j e c t s ' speech ( c f . L e h i s t e § Peterson, 1959, f o r d e t a i l s ) . I n i t i a l / s / plus stop consonant plus vowel c o n f i g u r a t i o n s can be segmented without d i f f i c u l t y by examination of the speech wave trace and the duplex o s c i l l o g r a m t r a c e . The f i r s t t r a c e , the speech s i g n a l , can be examined i n conjunction w i t h the negative amplitude of the duplex, or second, trace to de-termine an / s / ; ( c f . L e h i s t e and Peterson, 1959). The conso-nants /p/, / t / and /k/ immediately f o l l o w i n g the / s / can also be e a s i l y d i s t i n g u i s h e d , since the trace of the speech s i g n a l and duplex both f o l l o w the z e r o - l i n e . Vegetative sounds made by the s u b j e c t s , such as s n o r t s , breaths, l i p smacks or c l e a r -ing of the t h r o a t , made segmentation somewhat e a s i e r by adding n a t u r a l pauses between o f t e n i n d i s t i n c t utterances. The maximum e r r o r measurement which r e s u l t e d from the above procedure was ten m i l l i s e c o n d s , when one subject's data were measured again by the experimenter. This maximum e r r o r was present on only one of f i f t y measurements made. Utterances which presented problems i n segmentation (e.g., very r a p i d speech, or instances of dubious phonetic t r a n s c r i p -t i o n s ) were, as noted above, c l a s s i f i e d by examining wide-band spectrograms. CHAPTER 4 RESULTS 4.0 I n t r o d u c t i o n Using the "tongue-twisters", each of which contained twelve instances of e i t h e r /sp-/, / s t - / or /sk-/ i n word-i n i t i a l p o s i t i o n (as described i n Chapter 3 and given i n Appendix C), a sample of the s i x su b j e c t s ' speech under normal c o n v e r s a t i o n a l c o n d i t i o n s was obtained. Using the same "to n g u e - t w i s t e r s " produced at a s u b j e c t i v e l y f a s t e r r a t e of speech was a procedure determined to be e f f e c t i v e f o r generating speech e r r o r s i n the p i l o t s t u d i e s . A cor-pus of speech e r r o r s i n v o l v i n g the w o r d - i n i t i a l c l u s t e r s was al s o gathered and subsequently c l a s s i f i e d . The s i x subjects produced a t o t a l of 228 e r r o r s , which could be c a t e g o r i z e d i n t o s i x types, as f o l l o w s (see also Table 1): (1) Omission: d e l e t i o n of a phoneme or s y l l a b l e ; e.g., spotted -potted. (2) A d d i t i o n : epenthesis of a vowel or consonant; e.g., skimpy -+ kskimpy. (3) S u b s t i t u t i o n : phonetic replacement of d e v i a t i o n from the t a r g e t phoneme; e.g., s k i n skim. (4) Checked H e s i t a t i o n : i n s e r t i o n of g l o t t a l stop as a pausal phenomenon; e.g., s c a l l i o n s -> s c a l l ? i o n s . 44 45 (5) Prolonged H e s i t a t i o n : unusual lengthening or prolonga-t i o n of a phoneme; e.g., s t a l w a r t -* s t : a l w a r t . (6) R e p e t i t i o n : reproduction of a word, when the f i r s t p roduction i s h a l t e d f o l l o w i n g the f i r s t phoneme, c l u s t e r , s y l l a b l e or e n t i r e word; e.g., s p i r i t e d -*• s p s p i r i t e d . I n s e r t Tables 1 and 2 about here. More than f i f t y percent of a l l e r r o r s obtained were of type (6), i . e . , R e p e t i t i o n E r r o r s . These i n tu r n could be c l a s s i f i e d i n t o s i x sub-types, when grouped according to the repeated segment or segments, as f o l l o w s (see Table 2): (1) Phoneme R e p e t i t i o n : production of the i n i t i a l / s / , fol l o w e d by a pause and then production of the f u l l word; i . e . , / s * (pause) sC.../. (2) C l u s t e r R e p e t i t i o n w i t h a Pause: production w i t h a pause between the i n i t i a l c l u s t e r and r e p e t i t i o n of the e n t i r e word; i . e . , /s*C* (pause) sC.../. (3) C l u s t e r R e p e t i t i o n without a Pause: same as (2 ) , but without a pause between the e r r o r c l u s t e r and the repe-t i t i o n ; i . e . , /s*C* sC.../. (4) / s C V / - S y l l a b l e R e p e t i t i o n : production of an open s y l l a -b l e , f ollowed by a pause and the production of the whole word; i . e . , /s*C*V (pause) sC.../. TABLE 1. Types and Numbers of Speech Errors Produced by Each Subject. E r r o r Type Subj ect Omis-sion Addi-tion Substi-tution Checked Prolong Hesita- Hesita-tion tion Repeti-tion TOTAL ' 1 2 ' 3 4 5 6 4 8 1 5 7 2 2 1 1 1 2 4 2 1 2 1 1 6 1 2 5 11 3 19 17 1 18 28 15 • 33 11 13 32 52 . 24 59 37 24 TOTAL 25 6 10 13 56 118 228 TABLE 2. Types and Numbers of Repetition Errors Produced by Each Subject Subj ect Phoneme R e p e t i t i o n Cluster Cluster w/ Pause No Pause E r r o r Type Syllable Syllable sCV- sCVC-Entire Word TOTAL 1 2 3 4 5 6 2 9 5 3 4 3 2 10 7 3 2 2 4 1 1 5 5 2 8 5 4 4 8 1 5 3 1 1 3 3 2 18 28 15 33 11 13 TOTAL 19 26 13 29 22 9 118 46 47 (5) /sCVC/-Syllable Repetition.: production of a closed s y l l a b l e , f o l l o w e d by a pause and the production of the e n t i r e word; i . e . , /s*C*VC (pause) sC.../. (6) Word R e p e t i t i o n : production of the e n t i r e word followed by i t s r e p e t i t i o n . The above r e p e t i t i o n e r r o r types were examined w i t h respect to the d u r a t i o n of the f r i c a t i v e plus stop consonant c l u s t e r s , as w e l l as to that of the i n d i v i d u a l segments of which they were composed. 4.1 C o n t r o l Group Data A speech sample at each subject's normal r a t e of speech was obtained; and the durations of the /sC-/ c l u s t e r s , i n c l u d -ing i n t e r n a l segments, were then measured and analyzed. The d u r a t i o n data thus obtained provide norms f o r t h i s study and and h e r e a f t e r r e f e r r e d to as the Control Group Data (CGD). The summary s t a t i s t i c s f o r these data are presented i n Table 3. In s e r t Table 3 about here. The CGD were subjected to B a r t l e t t ' s t e s t f o r homogeneity of variance across s u b j e c t s : The chi-square values obtained d i d not exceed the c r i t i c a l value f o r s i g n i f i c a n c e (p > .01); i t was t h e r e f o r e assumed that the s u b j e c t s ' i n d i v i d u a l data could be pooled f o r purposes of f u r t h e r e v a l u a t i o n and analy-s i s . TABLE 3. C o n t r o l Group Data (Normal Conversational Rate) f o r W o r d - I n i t i a l Consonant C l u s t e r s /sp-/, / s t - / , /sk-/: Means and Standard Deviations of Segmental Durations ( i n ms). Segment Subject 2 3 4 5 6 Mean N=12 N=12 N=12 N=12 N=12 N=12 N=72 /s / X = 76.67 114.17* 140.00 89.17 91.67* 74.17 97.64 P sd= 17.23 53.17 32.75 31.47 31.86 • 21.09 39.56 /p/ X = 80.83 80.00* 90.83 82.50 91.67* 85.00 85.14 sd= 12.22 11-28 18.81 23.01 9-37 16.79 16.08 /sp/ X = 157.50 194.17* 230.83 171.67 183.33* 159.17 182.78 sd= 20.39 57.91. 46.41 49.70 34.20 23.14 46.90 /s / X = 100.83 117.50 123.33 87.50 80.00 93.33 100.42 sd= 15.64 21.79 28.39 28.32 20.45 27.08 28.01 ft/ X = 40.83 39.17 43.33 57.50 '44.17 45.00 45.00 sd= 18.32 10.84 11.55 8.66 13.79 17.84 14.73 / s t / X = 141.67 156.67 166.67 145.00 124.17 138.33 145.42 sd= 29.80 21.03 35.51 29.70 24.29 24.43 30.11 / s k / X = 93.33 107.50 125.00 85.00* 95.83 92.50 99.86 sd= 24.25 26.33 25.05 21.95 17.30 1.8.65 25.37 /k/ X = 57.50 63.33 61.67 61.67* 59.17 56.67 60.00 sd= 20.51 11.55 11.93 12.67 7.93 13.03 13.22 /sk/ X = 150.83 170.83 186.67 146.67* 155.00 149.17 159.86 sd= 43.16 29.38 28.39 28.71 21.53 29.68 33.04 * Due to subject error, N=ll for these entries; mean value added in each case to yield N=12 in order to standardize observations across a l l subjects. 48 49 Several analyses of variance were c a r r i e d out: A two-way c l a s s i f i c a t i o n demonstrated s i g n i f i c a n t d i f f e r e n c e s among s u b j e c t s , segments and the i n t e r a c t i o n of these (see Table 4); but wi t h respect to s u b j e c t - b y - c l u s t e r i n t e r a c t i o n , no such s i g n i f i c a n c e could be found (see Table 5). Under each of these analyses, a Newman-Keuls t e s t (with p < .05) was c a r r i e d out, which i n d i c a t e d that the i n d i v i d u a l c l u s t e r s comprised homogeneous subsets, as d i d each of the stop con-sonants, whereby a l l i n i t i a l /s/-segments f e l l i n t o the same subset (see Tables 4a and 5a). In s e r t Tables 4 - 7a about here. One-way analyses of variance were also c a r r i e d out wi t h the data from a l l subjects pooled. As expected, the segments and c l u s t e r s showed s i g n i f i c a n t d i f f e r e n c e s , and the Newman-Keuls t e s t (with p_ < .05) demonstrated the same homogeneous subset groupings as were found i n the two-way c l a s s i f i c a t i o n (see Tables 6 and 6a, 7 and 7a). A general i n t e r p r e t a t i o n of the CGD can be made' by con-s i d e r a t i o n of the o v e r a l l means (as given i n the l a s t column of Table 3 ), which are presented g r a p h i c a l l y i n Figure 2. The d u r a t i o n of / s / before any of the stop consonants was approximately 100 ms, wi t h / s / before /p/ being s l i g h t l y s h o r t e r . The mean d u r a t i o n of the stop consonants ranges from 45-85 ms, w i t h / t / having the s h o r t e s t and /p/ having TABLE 4. Summary of Analysis of Variance: Control Group Data --Cluster Segments and Subjects. Source of V a r i a t i o n Sum of Squares d.f. Mean Square F P Subjects (S) 29585.18 5 5917.04 12.566 <.0001 Segments (P) 198937.95 5 39787.59 84.497 . <.0001 S x P 42581.48 25 1703.26 3.617 <.0001 Er r o r 186466.65 396 470.88 TABLE 4a. Newman-Keuls Summary Table: Control Group Data ( p < .05) --Cluster Segments and Subjects. Homogeneous Subsets Subj ects C l u s t e r Segments (1) 6, 1, 5, 4 (1) /s /, /s /, /s / (2) 2 (2) / t / (3) 3 (3) /k/ (4) /p/ 50 TABLE 5. Summary of Analysis of Variance: Control Group Data --Clusters and Subjects. Source of Sum of A -F Mean V a r i a t i o n Squares U . X . Square r P Subjects (S) 59170.01 5 11834.07 10.412 < .0001 C l u s t e r s (C) 51112.03 2 25556.02 22.484 < .0001 S x C 13860.18 10 1386.02 1.219 > .25 E r r o r 225049.98 198 1136.62 . TABLE 5a. Newman-Keuls Summary Table: Control Group Data (p < .05) Clusters and Subjects. Homogeneous Subsets Subjects C l u s t e r s (1) 6, 1, 5,. 4 (1) / s t / (2) 2 (2) /sk/ (3) 3 (3) /sp/ 51 TABLE 6. Summary of Analysis of Variance: Control Group Data --Cluster Segments, One-Way Classification. Source of V a r i a t i o n Sum of Squares d.f. Mean Square F P Segments 198937.95 5 39787.59 65.535 < .0001 E r r o r 258633.31 426 607.12 TABLE 6a. Newman-Keuls Summary Table: Control Group Data ( p < .05 ) --Cluster Segments. Homogeneous Subsets: Segments (1) / s p / , / s k / , / s t / (2) / t / (3) /k/ (4) /p/ 52 TABLE 7. Summary of Analysis of Variance: Control Group Data -Clusters, One-Way Classification. Source of V a r i a t i o n Sum of , f Squares a , t ' Mean p Square — P C l u s t e r s 51112.03 2 25556.02 18.26 < .0001 E r r o r 298080.53 213 1399.44 TABLE. 7a. Newman-Keuls Summary Table: Clusters. Control Group Data (p < .05 ) — Homogeneous Subsets: C l u s t e r s (1) / s t / (2) /sk/ (3) /sp/ 53 54 the longest d u r a t i o n . The c l u s t e r s range i n mean d u r a t i o n from about 145 ms f o r / s t / to 183 ms f o r /sp/, w i t h /sk/ i n between at 160 ms. The d u r a t i o n of the c l u s t e r s i s c o n s i s -t e n t l y of the same order as wit h the i n d i v i d u a l stop seg-ments, due to the s i m i l a r i t y of the durations of the / s / -segments. • I n s e r t Figure 2 about here. 4.2 Experimental Group Data The s i x R e p e t i t i o n E r r o r c l a s s e s (hereafter c a l l e d Ex-perimental Groups #1 - #6) were examined se p a r a t e l y by subject f o r segments and c l u s t e r s , the summary s t a t i s t i c s f o r which are presented i n Table 8. Ins e r t Table 8 about, here. In Experimental Group #1 (Phoneme R e p e t i t i o n ) , / s / i s c o n s i s t e n t l y s h o r t e r the / s * / ( i . e . , the e r r o r p r o d u c t i o n ) , regardless of the stop consonant which f o l l o w s . The segment /p/ i s the longest stop consonant (ca. 100 ms), followed by /k/ (87.5 ms). and f i n a l l y by /%/ (50 ms); t h i s i s c o n s i s t e n t w i t h the f i n d i n g s f o r the CGD. The r e l a t i v e o rdering of the c l u s t e r durations f i n d s /sk/ to have the longest mean dura-t i o n and / s t / the s h o r t e s t ; t h i s d i f f e r s from the CGD i n that 200 D U R A T I O N ( i n ms ) 100 0 100 200 s | I t s i k s ! I s 1 K s I , J s I t s " fk s s s s s J " ] p ] p S . l S.2 S.3 S.4 S.5 S.6 MEAN FIGURE 2. C o n t r o l Group Data: Segment and C l u s t e r Duration. Mean values f o r each subject. 55 TABLE 8. Summary Statistics for Segmental and Cluster Durations (in ms), in Six Repetition Error Groups (termed Experimental Groups). ( Mean / Standard Deviation / Number of Observations ) Experimental * * s*t* t st s*k* , sk #1: s*: 136.67/ 66.58/ 3 127.50/ 92.15/ 4 162.50/154.17/12 Phoneme s: 90.00/ 10.00 107.50/ 57.37 154.17/122.14 Repetition C*: (s* - sC) C: 100.00/ 50.00 50.00/ 24.49 87.50/ 61.52 s*C*: sC: 190.00/ 45.83 157.50/ 71.36 241.67/136.90 #2: s*: 120.00/ 62.05/ 5 128.00/ 35.64/ 5 118.75/ 45.73/16 Cluster s: 68.00/ 16.43 100.00/ 40.00 116.25/ 31.60 Repetition C*: 208.00/110.77 318.00/261.29 229.37/227.29 (with Pause) C: 82.00/ 4.47 52.00/ 19.24 66.87/ 29.15 s*C*: 328.00/129.31 446.00/273.28 348.12/242.63 sC: 150.00/ 15.81 152.00/ 48.68 183.12/ 48.82 #3: s*: 128.00/ 48.68/ 5 140.00/ 56.57/ 2 158.33/105.53/ 6 Cluster s: 118.00/ 65.35 95.00/ 21.21 131.67/ 36.01 Repetition C*: 462.00/308.50 270.00/296.98 376.67/164.03 (No Pause) C: 45.00/ 25.50 40.00/ 28.28 65.00/ 10.49 s*C*: 590.00/543.29 410.00/240.42 535.00/231.58 sC: 163.00/ 85,41 135.00/ 7.07 196.67/ 41.79 #4: •s*: 118.33/ 51.15/ 6 87.69/ 33.20/13 103.00/ 32.68/10 sCV-Syllable s: 116.67/ 54.65 90.77/ 33.78 102.00/ 39.38 Repetition C*: 160.83/154.93 79.23/ 57.22, 92.00/ 41.04 C: 70.00/ 17.89 43.08/ 20.97 60.00/ 23.09 s*C*: 279.17/156.67 166.92/ 81.69 195.00/ 54.42 sC: 186.67/ .61.21 133.85/ 50.42 162.00/ 53.71 #5: s*: 116.67/ 46.19/ 3 80.00/ 19.27/ 8 121.82/ 47.08/11 sCVC-Syllable s: 120.00/ 50.00 105.00/ 51.27 105.45/ 26.22 Repetition C*: 123.33/ 77.67 .40.00/ .13.09 148.18/254.67 C: 90.00/ 26.46 41.25/ 16.42 45.45/ 13.68 s*C*: 240.00/122-. 88 120.00/ 22.04 270.00/286.23 sC: 210.00/- 72.11 146.25/ 55.79 150.91/ 35.34 #6:. s*: 115.00/ 49.50/ 2 60.00/ 0.00/ 2 108.00/ 32.71/ 5 Word s: 75.00/21.21 110.00/ 14.14 108.00/ 22.80 Repetition C*: 115.00/ 7.07 55.00/ 7.07 82.00/ 17.89 C: 80.00/ 28.28 80.00/ 28.28 70.00/10.00 s*C*: 230.00/ 56.57 115.00/7.07 190.00/47.43 sC: 155.00/ 49.50 190.00/ 14.14 178.00/ 30,35 56 57 /sk/ and /sp/ are reversed i n order with respect to d u r a t i o n . Because of the phonetic shape of t h i s experimental group, no segment /C*/ or c l u s t e r /s*C*/ e x i s t s (see a l s o Figure 4). In Experimental Group #2 ( C l u s t e r R e p e t i t i o n w i t h a Pause), the e r r o r productions / s * / , /C*/ and /s*C*/ are a l l longer i n d u r a t i o n than the c o r r e c t e d r e p e t i t i o n s / s / , /C/ and /sC/, r e s p e c t i v e l y . Considering the e r r o r c l u s t e r s : /s*t*/ i s the longest (446 ms) and /s*p*/ the s h o r t e s t (328 ms), a f i n d i n g which i s not c o n s i s t e n t w i t h the other experimental groups. The c l u s t e r /sk/ has the longest dura-t i o n (183 ms) and /sp/ the s h o r t e s t (150 ms), which i s not c o n s i s t e n t w i t h the CGB (see Figure 3). Experimental Group #3 ( C l u s t e r R e p e t i t i o n without a Pause) patterns a f t e r the above group, where /s*/, /C*/ and /s*C*/ are c o n s i s t e n t l y longer i n d u r a t i o n than / s / , /C/ and /sC/, r e s p e c t i v e l y . In t h i s group, /C*/ i s d e f i n i t i v e l y longer than in-any other experimental group: the c l u s t e r /s*p*/ i s the longest (590 ms) and / s * t * / the s h o r t e s t (410 ms), while /sk/ i s the longest (197 ms) and / s t / the s h o r t e s t (135 ms) of the c o r r e c t e d productions. The c l u s t e r /.st/ i s the only one which i s c o n s i s t e n t w i t h the CGD (see a l s o Figure 3). Experimental Group #4 . (/sCV/.-Syllable R e p e t i t i o n ) a l s o shows a l l e r r o r segments to be longer than non-error segments. Here, /s*p*/ and /sp/ are the longest and / s * t * / and / s t / the s h o r t e s t c l u s t e r s . 58 In Experimental Group #5 (/sCVC/-Syllable R e p e t i t i o n ) , the non-error segments and c l u s t e r s are longer than the e r r o r segments and c l u s t e r s , r e s p e c t i v e l y . This d i f f e r s from a l l other groups discussed thus f a r . Of the non-error productions, /s^/, /p/ and /sp/ have the longest d u r a t i o n s , while /s|:/> /k*/ and /s*k*/ show the longest durations of the e r r o r segments and c l u s t e r s (see Figure 3). Experimental Group #6 (Word R e p e t i t i o n ) demonstrates /s /, ft/ and / s t / to have the longest d u r a t i o n s , while /s /, z p /p/ and /sp/ have the s h o r t e s t . Of the e r r o r data, / s * / , /p*/ and /s*p*/ are the longest, while / s * / , / t * / and / s * t * / are the s h o r t e s t , thus c o n f l i c t i n g with the r e s u l t s f o r the CGD (see al s o Figure 4). ' Ins e r t Figures 3 and 4 about here. In s e r t Table 9 about here. In order to g e n e r a l i z e o b s e r v a t i o n s , the data from Ex-perimental Groups #2-#5 were pooled, the r e s u l t s of which are presented i n Table 9. This summary i l l u s t r a t e s that /s*p*/ at 368 ms and /sp/ at 174 ms e x h i b i t the longest c l u s t e r durat i o n s , while / s * t * / at 2 21 ms and / s t / at 141 ms FIGURE 3. Experimental Group's #2-#5: Segment and C l u s t e r Duration. Mean values f o r e r r o r and c o r r e c t productions. 59 D U R A T I O N ( i n ms ) 200 100 I 100 — I -200 — I — 300 — I — 400 —I E '.V.'.'.'.'.'.'.'.'.'.V. ( ( ' • • I I I , ] p EG#1 M y m i n t u n a r i f i 1 r r r y u I I / I t r i i i t w • » . . • . . . . . . „ . . < > . ",ir,i-n-itiiii r't,iii,ft.rt,in| r 1 r EG#6 FIGURE 4, Experimental Groups #1 and #6: Segment and C l u s t e r Duration. Mean values f o r e r r o r and co r r e c t productions. 60 TABLE 9. Summary of Repetition Error Data for Experimental Groups #2-#5: Segmental and Cluster Durations (in ms). (Mean / S.D. / N ) Segment E r r o r Production Segment Correct Production /s*/ 121,05 / 48.64 / 19 /s / 104.74 / 51.25 / 19 / s * / 96.43 / 36.54 / 28 /s / 96.79 / 38.50 / 28 / s * / 121.40 / 55.75 / 43 /s^/ 112.33 / 33.23 / 43 /p*/ 246.58 / 222.49 / 19 /p/ 69.74 / 24.41 / 19 / t * / 124.29 / 162.74 / 28 ft/ 43.93 / 19.12 / 28 /k*/ 197.21 / 213.81 / 43 /k/ 59.53 / 23.50 / 43 /s*p*/ 367.63 / 240.61 / 19 /sp/ 174.47 / 61.30 / 19 / s * t * / 220.71 / 180.96 / 28 / s t / 140.71 / 48.45 / 28 /s*k*/ 318.60 / 242.54 / 43 /sk/ 171.86 / 47.42 / 43 61 62 e x h i b i t the s h o r t e s t ; t h i s i s c o n s i s t e n t w i t h the CGD f i n d -ings. The segments / s | / at 121 ms and / s ^ / at 112 ms are only s l i g h t l y longer i n the l a t t e r case than / s * / at 121 ms and /s / at 105 ms, with /s*/ at 96 ms and /s^/ at 97 ms r e -maining the s h o r t e s t . Also c o n s i s t e n t w i t h the CGD i s the f i n d i n g that /p*/ at 247 ms and /p/ at 70 ms are the longest stop segments, .whereby It*I at 124 ms and / t / at 44 ms are the s h o r t e s t . The f o l l o w i n g general observations may be made: (1) Segments and c l u s t e r s i n the e r r o r , or f i r s t , productions are longer than t h e i r r e s p e c t i v e counterparts i n the c o r r e c t e d , or second, productions. (2) The sho r t e s t c l u s t e r s are / s * t * / and / s t / , which also c o n t a i n the sh o r t e s t segments / s * / , / t * / , /s / and / t / . (3) The longest /s/-segments are /s£/ and / s ^ / * (4) The longest stop consonants are /p*/ and /p/. (5) The longest c l u s t e r s are /s*p*/ and /sp/. I f we consider the r e s u l t s of o v e r a l l means obtained i n the c o r r e c t productions i n l i g h t of the CGD f i n d i n g s , we note t h a t : (1) The d u r a t i o n of / s / before any stop consonant was approx-imately 100 ms i n the CGD and 104 ms i n the Experimental Group Data (EGD). (2) The d u r a t i o n of the stop consonant was 63 ms i n the CGD and 53 ms i n the EGD. (3) The d u r a t i o n of the c l u s t e r was 163 ms f o r both CGD and EGD. 63 The ranking of the segments and c l u s t e r s w i t h i n t h e i r re-s p e c t i v e groups also remains about the same. An i n t e r e s t i n g observation might be made regarding the s i m i l a r i t y of these means: Under c o n t r o l c o n d i t i o n s , subjects were speaking at a normal r a t e , while the same s u b j e c t s , when speaking under experimental c o n d i t i o n s , were speaking as f a s t as they could. This should, i t would seem, make the durations of the EGD segments and c l u s t e r s s h o r t e r than those of the CGD; but such was not the case. I t i s reasonable to suppose from these r e s u l t s that once an e r r o r i n a f i r s t p roduction i s made, perhaps the r a t e returns to the normal c o n v e r s a t i o n a l r a t e u n t i l the subject can once again p i c k up speed. Summary s t a t i s t i c s were a l s o d e r i v e d f o r the" pause, or delay time, between e r r o r and c o r r e c t productions. These s t a t i s t i c s show that the pause can be e l i m i n a t e d , but that a lengthening of the consonant"before the pause, when i t occurs, can have a mean value as great as 967 ms (such as i n Phoneme R e p e t i t i o n ) . Disregarding the cases where no pause occurs, the pause can be as short as 20 ms or as long as 6550 ms (with l i t t l e or no lengthening of the segment before i t ) . CHAPTER 5 DISCUSSION 5.0 General Considerations The i n t e n t of the present study was to examine system-a t i c a l l y s p e c i f i c aspects of speech e r r o r s through studying r e l a t i o n s h i p s of word i n i t i a l f r i c a t i v e plus stop consonant c l u s t e r s . The categories i n t o which e r r o r utterances are c l a s s i f i e d i n the l i t e r a t u r e p e r t a i n i n g to speech e r r o r s ( s p e c i f i c a l l y Boomer § Laver, 1968) were not adequate to des-c r i b e e r r o r s produced by subjects i n t h i s study. In a d d i t i o n to m i sordering, omission and replacement of segments ( c f . Boomer $ Laver, 1968), a d d i t i o n , h e s i t a t i o n and r e p e t i t i o n e r r o r s were produced. These l a t t e r c a tegories were found by c o n s u l t i n g the l i t e r a t u r e on delayed auditory feedback (DAF) ami wert adapted from the categories set up by Fairbanks and Guttman (1958) , since they proved to be the most a p p l i c a b l e to the present study. Approximately f i f t y percent of a l l e r r o r s produced by-subjects i n t h i s study were of the r e p e t i t i o n type. Categor-i z a t i o n of ..these e r r o r s was based on phonetic t r a n s c r i p t i o n s c a r r i e d out by the experimenter and a t r a i n e d p h o n e t i c i a n and supplemented by spectrographic a n a l y s i s where necessary ( f o r a l i s t of the e r r o r s and the phonetic t r a n s c r i p t i o n s , see Appendix D). This y i e l d e d s i x c a t e g o r i e s of r e p e t i t i o n e r r o r s based on t h e i r phonetic forms. 64 65 Speculation as to why r e p e t i t i o n e r r o r s are c o r r e c t e d and e r r o r s i n the other categories are not l e d to examination of the f i r s t productions f o r phonetic deviancies v i s - a - v i s the subsequent c o r r e c t i o n productions. I t was found that only about one t h i r d of the r e p e t i t i o n e r r o r s could be con-s i d e r e d as c o r r e c t i o n s because of phonetic a b n o r m a l i t i e s i n the f i r s t , or e r r o r , production. The phenomenon of excessive length associated w i t h the i n i t i a l c l u s t e r and i t s component segments was o f t e n noted, but i t was not considered as a phonetic abnormality. I t was observed t h a t the f i r s t phoneme or c l u s t e r i n the e r r o r s y l l a b l e or word was somewhat longer i n duration than might have been s u b j e c t i v e l y expected. This observation l e d i n turn to comparison of the durations of the i n i t i a l c l u s t e r s and t h e i r segments i n the e r r o r production w i t h the immediately f o l l o w i n g c l u s t e r p r o d u c t i o n , or c o r r e c -t i o n . In order to measure o b j e c t i v e l y and compare these dura-t i o n s , o s c i l l o g r a m s were produced-and measured f o r each sub-j e c t ' s normal production of the " t o n g u e - t w i s t e r s " , then f o r each subject's e r r o r s which occurred i n subsequent r a p i d pro-ductions. The r e s u l t s obtained were subjected to s t a t i s t i c a l a n a l y s i s (as d e t a i l e d i n the previous chapter). Findings s p e c i f i c to the c o n t r o l group data w i l l be discussed f i r s t . 5.1 Discussion of the C o n t r o l Group The a n a l y s i s of variance c a r r i e d out on the c o n t r o l group data showed s i g n i f i c a n t d i f f e r e n c e s between subjects and be-tween phonemes, as w e l l as c l u s t e r s (cf. Tables 4-7). The 66 Newman-Keuls t e s t grouped four of the subjects i n t o one homogeneous subset, while Subject 2 and Subject 3 were each separately grouped. The segregation of these two subjects may have r e s u l t e d from the f a c t that both spoke w i t h sub-j e c t i v e l y more p r e c i s e a r t i c u l a t i o n and somewhat more slow l y than the other subjects (but also d i f f e r e n t l y enough from one another to be grouped s e p a r a t e l y ) . Such f a c t o r s would tend to lengthen segments and c l u s t e r s i n the speech of these two subjects and thus account f o r the d i f f e r e n c e s i n the Newman-Keuls t e s t ( cf. Tables 4a and 5a). The four s i g n i f i c a n t subsets f o r segments e x h i b i t e d by the Newman-Keuls t e s t ( cf. Tables 4a and 6a) and the three subsets f o r c l u s t e r s ( cf. Tables 5a and 7a) segregate (a) a l l /s/-segments as a group from /p/, ft/ and /k/, each of which are also grouped s e p a r a t e l y , and (b) the c l u s t e r s , / s t / from /sk/ from ./sp/. Examination of the means ( c f . Table 3) i n d i c a t e s that / s / before /p/ i s s l i g h t l y s h o r t e r than / s / before / t / . o r /k/, wi t h / t / being the s h o r t e s t stop and /p/ the l o n g e s t , c o n s i s t e n t w i t h the data reported by Schwartz (1970). Based on the means and the Newman-Keuls t e s t group-ings i n t h i s study, i t i s reasonable to speculate t h a t , s i n c e duration of the s i b i l a n t i s s i m i l a r i n each context, i t i s the stop consonant which u l t i m a t e l y determines the du r a t i o n of the c l u s t e r as a whole. . Indeed, when one examines the means : and the subset orderings f o r the stop consonants and f o r the c l u s t e r s , the same r e l a t i o n s h i p h olds; i . e . , /p/ i s . l o n g e r than ,/k/, which i s longer than / t / , and /sp/ i s longer than /sk/, which i s longer than / s t / . 67 The above f i n d i n g s are i n agreement wi t h those o f Borden and Gay (1975) w i t h respect to the groupings and re-l a t i v e orderings of segments comprising w o r d - i n i t i a l c l u s t e r s ; due perhaps .to t h e i r s u b j e c t s ' producing words i n i s o l a t i o n , t h e i r values f o r the durations of these segments are somewhat l a r g e r i n a l l cases. I n c o n s i s t e n t with t h e i r f i n d i n g s i s the r e l a t i v e o r d e r i n g of c l u s t e r l e n g t h s , s i n c e t h e i r data i n d i -cate that /sk/ i s s l i g h t l y longer than / s t / , which i s longer than /sp/. This suggests that f o r t h e i r three subjects the s i b i l a n t may be a greater determiner of c l u s t e r l e n g t h , which i s i n contrast w i t h the r e s u l t s of the present study. 5.2 D i s c u s s i o n of the Experimental Groups In experimental groups #l-#6 the d u r a t i o n of the f i r s t , or e r r o r , production was contrasted w i t h the second, or cor-r e c t e d , production. Because of the d i f f e r e n c e s i n rate of speech and other u n c o n t r o l l e d v a r i a b l e s , these speech e r r o r data cannot be l e g i t i m a t e l y compared w i t h the c o n t r o l group f i n d i n g s ; however, the general trends i n the two groups of data can be compared i n order to a s c e r t a i n whether s i m i l a r i t i e s e x i s t . Experimental groups #2-#5 were combined i n order to ob-serve more general trends i n the data. Experimental group #1 was not used, since i t contained no e r r o r c l u s t e r ; and experimental group #6 was not considered because i t i n v o l v e d r e p e t i t i o n of a whole word r a t h e r than of an i n i t i a l c l u s t e r 68 or s y l l a b l e . Groups #1 and #6 a l s o contained s.mall numbers of observations and were thus l e s s l i k e l y to a f f e c t group trends. As mentioned p r e v i o u s l y , subjects were t o l d t o produce the experimental utterances at t h e i r f a s t e s t speaking r a t e . Assuming that experimental utterances were indeed produced at a " f a s t e s t " rate of speech, one might f u r t h e r assume that segmental and c l u s t e r length would decrease; however, t h i s i s not the case, as can be seen i n Tables 3 and 8. One ex-p l a n a t i o n f o r t h i s could be that only vowels, resonants and perhaps pauses make a d i f f e r e n c e to r a t e , w h i l e stops and f r i c a t i v e s are only s l i g h t l y a f f e c t e d , i f at a l l . In addi-t i o n , a p o s s i b l e explanation f o r t h i s s u b j e c t i v e l y f a s t e r rate of speech could be the change i n d u r a t i o n of the h e s i t a -t i o n pause and the r e l a t i o n of semantic content to pausal time. As Goldman-Eisler (1968) r e p o r t s , v a r i a t i o n s i n the o v e r a l l rate of speech or an increase i n rate were found to be v a r i a t i o n s i n the amount of pausing. She concludes that rate of speech based s o l e l y on a r t i c u l a t o r y a c t i v i t y remained r e l a t i v e l y i n v a r i a n t . Goldman-Eisler (1968) also observes trends i n pausal time i n r e l a t i o n to semantic content. When subjects i n t e r p r e t e d meaning, pausal time was twice as great as when they described content. She examined t h i s phenomenon w i t h respect to degrees 69 of spontaneity i n speech. Where semantic content becomes l e s s and l e s s a f a c t o r i n speech, as i n r e p e t i t i o n of the same utterance, she found that there was a d e c l i n e i n pausal time a f t e r the f i r s t r e p e t i t i o n and a f u r t h e r decrease i n subsequent r e p e t i t i o n s . Considering these i n t e r p r e t a t i o n s , one would not assume that c l u s t e r length would decrease w i t h an increase i n . r a t e but would remain approximately the same, a t t a i n i n g i t s short-est d u r a t i o n i n the l a s t ( i . e . , f i f t e e n t h ) r e p e t i t i o n where semantic content i s most f a m i l i a r . General r e s u l t s can be grouped f o r d i s c u s s i o n purposes: The longest s i b i l a n t s are those before the v e l a r /k/, while the longest stop consonant i s the b i l a b i a l /p/ and the short-est the a l v e o l a r ft/. In the c o n t r o l group data, the stop consonant xvas considered to determine the length of the c l u s t e r . Hypothesizing that such a c o n s t r a i n t holds f o r the experimental group data, one might expect the c o r r e c t e d /sp/-c l u s t e r s to be the longest and the c o r r e c t e d / s t / - c l u s t e r s to be the s h o r t e s t . The r e s u l t s presented i n Table 9 support such an hypothesis. A p o s s i b l e p h y s i o l o g i c a l e x p l a n a t i o n f o r the f i n d i n g that /sp/ and /sk/ are longer i n d u r a t i o n than / s t / would be that the former i n v o l v e slower moving a r t i c u l a t o r y musculature (e.g., l i p s and body of the tongue) f o r the stop p r o d u c t i o n , while both / s / and / t / " i n v o l v e f a s t e r moving, more h i g h l y i n -nervated tongue t i p musculature. Furthermore, / t / . i s homorganic 70 with /s/ ( i . e . , place of a r t i c u l a t i o n i s the same, only the manner d i f f e r s ) , whereas the heterorganic c l u s t e r s /sp/ and /sk/ r e q u i r e s e v e r a l d i f f e r e n t muscles i n order to complete the a r t i c u l a t o r y gestures. These d i f f e r e n t movements should tend ( l o g i c a l l y ) to make production slower f o r /sp/ and /sk/ than f o r / s t / . Haggard (1973) a l s o discusses a b b r e v i a t i o n i n c e r t a i n homorganic c l u s t e r s : He r e p o r t s v a r i e d i n d i v i d u a l d i f f e r e n c e s and supposes that durations can be c o n t r o l l e d by o r a l pressure feedback. Perhaps the most i n t r i g u i n g f i n d i n g of the present i n -v e s t i g a t i o n i s that the elements of an e r r o r c l u s t e r are a l -ways longer than those of the r e p e t i t i o n , or c o r r e c t i o n pro-duction. I t i s of i n t e r e s t to note the p o s s i b l e e f f e c t s of s t r e s s and f o l l o w i n g vowel environment on c l u s t e r d u r a t i o n . Of the t h i r t y - s i x t e s t words embedded i n the three "tongue-twisters", a l l but two had primary s t r e s s on the f i r s t s y l l a b l e c o n t a i n i n g the c l u s t e r . Word s t r e s s was t h e r e f o r e not considered to be a major f a c t o r i n determining, c l u s t e r length.. S i m i l a r l y , the f o l l o w i n g vowel environment was analyzed to determine whether r e p e t i t i o n e r r o r s occurred more fre q u e n t l y before some vowels and not others. Of 118 r e p e t i -t i o n e r r o r s produced by the s i x s u b j e c t s , 39 e r r o r s occurred before the vowel / a e / , 29 before / 1 /, 24 before /a/, 9 before /e/, 5 before /oo/, 5 before / i /, 4 before /o/, and 3 before •/eL /. • • More r e p e t i t i o n e r r o r s occurred before the vowels /es/, l\l and /a/ than before the others. Since data were not analyzed to take f o l l o w i n g vowels i n t o account, no e x p l a n a t i o n 71 as to t h e i r s i g n i f i c a n c e can be o f f e r e d at t h i s time. The import of these r e s u l t s are the t o p i c of the next s e c t i o n . 5.3 T h e o r e t i c a l Considerations Results from the experimental group data s t r o n g l y sug-gest that d u r a t i o n of a segment or c l u s t e r already produced can a f f e c t subsequent a r t i c u l a t i o n s . I t seems that excessive d u r a t i o n of the c l u s t e r as a whole or of e i t h e r of i t s com-ponent parts may v i o l a t e some s o r t of timing c o n s t r a i n t on the system i n the production of a given u t t e r a n c e ; t h i s v i o l a -t i o n causes the production to be h a l t e d i n mid-word, a re-c a l i b r a t i o n to be e f f e c t e d , and a c o r r e c t i o n to be produced. I f , however, an e r r o r i s made w i t h respect to the phon-e t i c form.of the utterance (such as a s u b s t i t u t i o n e r r o r ) , the timing c o n s t r a i n t hypothesized above may not be v i o l a t e d , and the utterance would not have to be repeated. At present i t i s not known whether a d d i t i o n or omission e r r o r s v i o l a t e such a timing c o n s t r a i n t i n words, or i f other segments w i t h i n words lengthen or shorten to make room f o r an e x t r a element or to f i l l up an empty space. I f such a view i s tenable, the h e s i t a t i o n e r r o r s (where an intra-word pause or p r o l o n g a t i o n of a segment occurs) can be considered to be a step e a r l i e r than a r e p e t i t i o n e r r o r ; i . e . , w i t h i n c e r t a i n l i m i t s , seg-mental p r o l o n g a t i o n or pause i n s e r t i o n w i l l be enough f o r the system to r e c a l i b r a t e , but i f such r e c a l i b r a t i o n does not take place q u i c k l y enough, then production i s completely 72 h a l t e d , with the utterance being reproduced, y i e l d i n g what has been termed here a r e p e t i t i o n e r r o r . C onsideration s o l e l y of the r e p e t i t i o n e r r o r s does not allow f o r determination of whether (a) a delay i n production of the next phoneme caused the r e p e t i t i o n , (b) the r e p e t i t i o n i s a c o r r e c t i o n of an e x c e s s i v e l y long segment, or (c) the r e p e t i t i o n i s a c o r r e c t i o n of an e x c e s s i v e l y long c l u s t e r j u s t u t t e r e d . A combination of ( a ) , (b) and (c) i s a l i k e l y s o l u -t i o n ; i . e . , e x c e s s i v e l y long segments i n a c l u s t e r are pro-duced which causes a delay i n pr o d u c t i o n of the next phoneme, and the second production i n a r e p e t i t i o n e r r o r i s a c o r r e c t i o n of the timing v i o l a t i o n . In support of t h i s s o l u t i o n , l e t us again consider h e s i t a t i o n e r r o r s : As mentioned p r e v i o u s l y , no r e p e t i t i o n occurs i n such e r r o r s (perhaps) because (a) the duration of the h e s i t a t i o n i s not long enough to v i o l a t e a timing c o n s t r a i n t , and thus (b) the phoneme t r a n s i t i o n s have not been l o s t so that production of the word can be continued. This suggests that when a r e p e t i t i o n takes p l a c e , i t i s due to excessive duration which causes a delay v i o l a t i n g a timing con-, s t r a i n t and r e s u l t s i n l o s s of phoneme t r a n s i t i o n s . Conse-quently, the utterance cannot be continued, and a c o r r e c t i o n of the excessive d u r a t i o n o f the f i r s t p roduction i s ordered. I f such a n o t i o n i s adopted, then how might the system deter-mine that a timing c o n s t r a i n t has been v i o l a t e d ' a n d a c o r r e c t i o n c a l l e d for? 73 In order to accomplish c o r r e c t i o n of an element pre-v i o u s l y u t t e r e d , the system must f i r s t know that the element was i n c o r r e c t and what the t a r g e t should have been; i . e . , feedback must be present i n order f o r the system to recognize the d uration e r r o r , and comparison of output w i t h the o r i g i -n a l l y planned t a r g e t element must occur i n order f o r a correc-t i o n to be executed. The " a n a l y s i s - b y - s y n t h e s i s " model of speech p e r c e p t i o n and production (cf. B e l l et a l . , 1961) incorporates a feed-back method of speech spectrum a n a l y s i s by which c o r r e c t i o n of an element p r e v i o u s l y u t t e r e d could be accomplished. In terms of t h i s model, the "spectrum generator" produces output comparable to s t o r e d speech data ( i . e . , the f i r s t p r o duction of the u t t e r a n c e ) . The "comparator" then computes the d i f f e r -ence between the input speech s p e c t r a i t has r e c e i v e d v i a a feedback loop and the o r i g i n a l t a r g e t utterance j u s t generated. T r i a l s pectra are synthesized by the " s t r a t e g y component" u n t i l minimum e r r o r i s obtained i n matching and i n t h i s case a c o r r e c t i o n of a previous e r r o r utterance i s generated. On the p h y s i o l o g i c a l l e v e l , feedback can occur v i a the a c o u s t i c and/or p r o p r i o c e p t i v e channels (as discussed i n Sec-t i o n s 1.31 and 1.3.2), i . e . , v i a bone and a i r conduction and/or the gamma motor system. In the present study there i s no way to determine i f both channels are i n use at a l l times, or i f there i s i n t e r m i t t e n t monitoring by one or both channels dur-ing speech (as suggested e a r l i e r ) . One argument f o r i n t e r -m i t t e n t feedback during speech i s that omission and a d d i t i o n 74 e r r o r s are u s u a l l y not c o r r e c t e d . One may speculate that such e r r o r s do not v i o l a t e a timing c o n s t r a i n t placed on the word, but i t may be more l i k e l y t h a t such e r r o r s have not been "caught" by the system due to i n t e r m i t t e n t monitoring. In any case, the speaker/hearer receives feedback, and some comparison w i t h the o r i g i n a l t a r g e t must take place -- f a c t s which models of speech production (and perception) should account f o r . The notions discussed thus f a r may be viewed i n l i g h t of Fairbanks's (1954) i n t e r p r e t a t i o n of the speech production system as a servosystem. He suggested that there i s continu-ous monitoring v i a the a c o u s t i c mode, by which we compare out-put to input and thereby manipulate production. M o n i t o r i n g s o l e l y by means of the a u d i t o r y channel i s most u n l i k e l y , since an a d v e n t i t i o u s l y deafened person does not l o s e h i s speech immediately a f t e r an i n j u r y (as noted i n S e c t i o n 1.62) ; perhaps such an i n d i v i d u a l can r e l y on h i s p r o p r i o c e p t i v e feed-back from the a r t i c u l a t o r s to supplement b a r e l y d i s c e r n i b l e auditory s i g n a l s . For the normal hearing person, Abbs (19 73) proposed a " v a r i a b l e " ser>/:osystem ( c f . S e c t i o n 1.63), a model which c l o s e l y acquaints p r o p r i o c e p t i v e feedback and input/output comparisons. This speech production system i s e f f i c i e n t , s i n c e i t employs feedback depending on i t s requirements, v i a the gamma or s p i n d l e motor system. In order to consider which system might be i n o p e r a t i o n , l e t us speculate how r e p e t i t i o n e r r o r s might v i o l a t e timing c o n s t r a i n t s . 75 The v i o l a t i o n of a t i m i n g c o n s t r a i n t could be due to a "l a p s e " i n continuous a u d i t o r y feedback because of a f a s t e r rate of speech; i . e . , the a u d i t o r y monitoring system (and "comparator") lag behind production to the extent that a lengthening of segments occurs which, i n the case of r e p e t i t i o n e r r o r s , v i o l a t e s a timing c o n s t r a i n t . Perhaps f o r t h i s reason these e r r o r s c l o s e l y resemble the r e p e t i t i o n e r r o r s produced under the i n f l u e n c e of delayed auditory feedback, where a delay r e s u l t s i n such a " l a p s e " and a r e p e t i t i o n i s produced (cf. Fairbanks § Guttman, 1958; Lee, 1951). I f such a lapse does occur, then r e p e t i t i o n e r r o r s are not j u s t a production problem, but also a p e r c e p t i o n problem. In r e l a t i n g t h i s n o t i o n of auditory feedback's lagging speech p r o d u c t i o n to the " a n a l y s i s - b y - s y n t h e s i s " model, we might speculate that i t i s the "comparator", i n computing the d i f f e r e n c e between input and output, which has caused the delay, and a . r e c a l i b r a t i o n i s necessary f o r i t to catch up to production. The delay: thus v i o l a t e s a timing c o n s t r a i n t here too, and a c o r r e c t e d utterance must be produced. To account f o r the f a c t that not a l l speech e r r o r s are c o r r e c t e d , one may speculate that auditory feedback monitors only general sound patterns and i n t o n a t i o n , rhythm and s t r e s s p a t t e r n s , while p r o p r i o c e p t i v e feedback monitors i n t e r m i t t e n t l y f o r phonetic deviancies. In accordance w i t h Abbs (1973) t h i s i n t e r m i t t e n t p r o p r i o c e p t i v e feedback concerns the gamma or s p i n d l e motor system, which (1) maintains length or r a t e of 76 change of length of a muscle, (2) i n i t i a t e s c o n t r a c t i o n , (3) provides damping movements to prevent overshoot, and (4) supplements auditory feedback when necessary. There i s a reasonable b a s i s f o r supposing t h a t the gamma motor system i s o p e r a t i o n a l w i t h respect t o p h o n e t i c a l l y de-vi a n t r e p e t i t i o n e r r o r s (such as those observed i n t h i s study). I f we consider the s h o r t e s t pause between r e p e t i t i o n s observed, here approximately 20 ms, and the " t u r n around time" or delay f o r o p e r a t i o n o f the s p i n d l e system, ca. 20-80 ms, we f i n d that the two f i g u r e s o v e r l a p , and the i n t e r p r e t a t i o n i s not contra-d i c t e d . A l l t o l d , a model of speech production must account f o r normal u t t e r a n c e s , as w e l l as f o r speech e r r o r s . Moreover, i t must allow f o r appropriate types of feedback and be an e f f i c i e n t system f o r r e l a t i n g speech production to p e r c e p t i o n . The v a r i a b l e servomonitor system o u t l i n e d above incorporates both continuous auditory feedback and i n t e r m i t t e n t p r o p r i o -ceptive feedback, which are used i n p e r c e i v i n g input and using i t to manipulate output. This system a l s o provides a p l a u s i b l e account of the speech e r r o r s and of t h e i r production as des-c r i b e d i n t h i s study. T h e o r e t i c a l l y , the system advocated here provides an e f f i c i e n t means fo r producing, monitoring and c o r r e c t i n g speech production. 5.4 L i m i t a t i o n s of the Present Study One c f the purposes of t h i s i n v e s t i g a t i o n was :to examine speech e r r o r s under c o n d i t i o n s of r a p i d r e p e t i t i o n of three 77 "tongue-twisters". I t was hypothesized that e r r o r s obtained using a f a s t e r than normal speaking rate might be due to at l e a s t the f o l l o w i n g : (1) The words were s i m i l a r i n p h o n o l o g i c a l form and phonetic content ( i . e . , twelve occurrences each of /sp/, / s t / and /sk/ i n w o r d - i n i t i a l p o s i t i o n ) , and (2) the f a s t e r rate of speech which, when combined w i t h ( 1 ) , r e s u l t s i n e r r o r productions. These i n v o l v e v a r i a b l e s such as psycho-l o g i c a l and p h y s i c a l s t r e s s and f a t i g u e which were not con-t r o l l e d f o r -- i f indeed they can be c o n t r o l l e d f o r -- i n t h i s study. The f o l l o w i n g l i m i t a t i o n s apply to any i n t e r p r e t a t i o n of the data: (1) Generation, of speech e r r o r s may not produce the same type of e r r o r s as. those produced i n spontaneous speech. They may be due to memory l i m i t a t i o n s , which, when combined w i t h speak-ing r a t e , put undue s t r e s s on the speaker, who may then pro-duce "unnatural" e r r o r s . As such, the e r r o r s described may be a r t i f a c t s of the experimental method employed. (2) Words w i t h i n i t i a l f r i c a t i v e plus stop consonant c l u s t e r s were not c o n t r o l l e d f o r number of s y l l a b l e s , word c l a s s , s t r e s s placement or place i n sentence. The above were considered to be the major l i m i t a t i o n s of t h i s study, and due c o n s i d e r a t i o n f o r the c o n t r o l of such v a r i a b l e s should be given to fut u r e research i n t h i s area. 5.5 Summary and Conclusions The present study f i r s t examined the means by which the generation and the c l a s s i f i c a t i o n of speech e r r o r s could be 78 accomplished. I t was found that a " t o n g u e - t w i s t e r " , which contained many occurrences of words with' s i m i l a r p h o n o l o g i c a l form ( i . e . , w i t h w o r d - i n i t i a l /sp/, / s t / and /sk/ c l u s t e r s ) , produced at a subject's f a s t e s t speaking rate and repeated many times would generate the most speech e r r o r s . These e r r o r s were then c l a s s i f i e d according to categories combined from the l i t e r a t u r e on speech e r r o r s and on DAF research. F i f t y perr cent of a l l e r r o r s produced by subjects i n t h i s i n v e s t i g a t i o n were of the r e p e t i t i o n type. C o n s i d e r a t i o n of the p o s s i b l e causes of the e r r o r s encountered l e d to the d e t a i l e d examina-t i o n of the w o r d - i n i t i a l c l u s t e r and component segment dura-t i o n s . The experimental i n v e s t i g a t i o n y i e l d e d the f o l l o w i n g general r e s u l t s : (1) The stop consonant i n a given c l u s t e r seems to determine the o v e r a l l c l u s t e r d u r a t i o n , s i n c e the d u r a t i o n of / s / i r r e -s p e c t i v e of context remains f a i r l y constant. (2) The c l u s t e r s /sp/ and /sk/ are longer i n d u r a t i o n than / s t / , which r.iay be a t t r i b u t a b l e to the slower moving a r t i c u l a -tory musculature a s s o c i a t e d w i t h /p/ and /k/ production com-pared w i t h the f a s t e r moving, more h i g h l y i n n e r v a t e d tongue t i p musculature i n v o l v e d i n the.production of / s / and ft/. (3) The c l u s t e r segments i n the e r r o r productions were con-s i s t e n t l y longer i n d u r a t i o n than i n the second, or c o r r e c t e d , production (which approximated more c l o s e l y the values obtained f o r the c o n t r o l group data than might have been otherwise ex-pected due to methodological d i f f e r e n c e s ) .: 79 In l i g h t of the above r e s u l t s , i t was speculated that the excessive duration of the c l u s t e r (or of i t s component pa r t s ) v i o l a t e d a timing c o n s t r a i n t on the production of an utterance, whereupon phoneme t r a n s i t i o n s are l o s t to the sys-tem, f o l l o w i n g which a r e c a l i b r a t i o n must take place and a c o r r e c t i o n produced. From such c o n s i d e r a t i o n s i t was i n f e r r e d that feedback must be present i n order f o r the system to re-cognize the duration e r r o r , to compare i t w i t h planned output, and f i n a l l y to execute a c o r r e c t i o n . On the p h y s i o l o g i c a l l e v e l , feedback was considered to be both continuous (auditory channel) and i n t e r m i t t e n t (pro-p r i o c e p t i v e channel, i n v o l v i n g the gamma motor system); the l a t t e r may supplement auditory feedback and scan f o r deviant phonetic elements, w h i l e the former monitors general sound p a t t e r n s , p a r t i c u l a r l y suprasegmental p a t t e r n s . As a r e s u l t of these c o n s i d e r a t i o n s , i t was hypothesized that a timing c o n s t r a i n t i s imposed by the system. When t h i s c o n s t r a i n t i s v i o l a t e d , perhaps due to a delay i n aud i t o r y feedback processing occasioned by the f a s t e r than normal speak-ing r a t e , a speech e r r o r occurs. Because of t h i s , r e p e t i t i o n errors were regarded as much a perception problem as a produc-t i o n problem. REFERENCES Abbs, J.H. 1973. The influence of the gamma motor system on jaw movements during speech: A theoretical framework and some preliminary observations. Journal of Speech  and Hearing Research 16. 175-200. Alston, W., R.W. Angel, F.S. Fink, and W.W. Hofmann. 1967. Motor activity following the silent period in human muscle. Journal of Physiology 190. 189-202. [Cited in Abbs, 1973.] Bell, C.G., H. Fujisaki, J.M. Heinz, K.N. Stevens, and A.S. House. 1961. Reduction of speech spectra by analysis-by-synthesis techniques. Journal of the  Acoustical Society of America 33. 1725-1736. Bernstein,. N.A. 1967. The Coordination and Regulation of  Movements. Oxford: Pergamon Press. [Cited i n Ohala, 1973.] Berry, M., and J. Eisenson. 1956. Speech Disorders: Princi-ples and Practices of Therapy. New York: Appleton-Century-Crofts. Black, J.W. 1951. The effect of delayed side-tone upon vocal rate and intensity. Journal of Speech and Hearing  Disorders 16. 56-60. Boomer, D.S., and J.D.M. Laver. 1968. Slips of the tongue. British Journal of Disorders of Communication 3. 1.12. Borden, G.J..and T. Gay. 1975. Durations of articulator movements for /s/.-stop. clusters. Haskins Laboratories  Status Report on Speech Research SR-44. 147-160. Bowman, J.P. 1971. The Muscle Spindle and Neural Control of  the Tongue. Springfield, 111.: Charles C Thomas. Campbell, E.J.M. 1968. The respiratory muscles. Sound Pro-duction in Man, ed. by A. Bouhuys, .135-140. New York: . New York Academy of Sciences. [Cited in Abbs, 1973.J 80 81 Chase, R.A. 1958. Effect of delayed auditory feedback on the repetition of speech sounds. Journal of Speech and  Hearing Disorders 23. 583-590. Euler, C. von. 1966. Proprioceptive control in respiration. Muscular Afferents and Motor Control, ed. by R. Granit, 197-208. New York: Wiley. [Cited in Abbs, 1973.] Fairbanks, G. 1954. Systematic research in experimental phonetics: 1. A theory of the speech mechanism as a servosystem. Journal of Speech and Hearing Disorders 19. 133-139. Fairbanks, G. 1960. Voice and Articulation Drillbook, 2nd edition. New York: Harper & Row. Fairbanks, G., and N. Guttman. 1958. Effects of delayed auditory feedback upon articulation. Journal of Speech  and Hearing Research 1. 12-22. Freud, S. 1924. Zur Psychopathologie des Alltaglebens, 10th edition. Leipzig: Internationaler Psychoanalytischer Verlag. [Cited in Fromkin, 1971.] Fromkin, V.A. 1968. Speculations on performance models. Journal of Linguistics 4. 47-68. Fromkin, V.A. 1971. The non-anomalous nature of anomalous utterances. Language 47. -27-52. Fromkin, V.A; 1973. Slips of the tongue. Scientific Amer-ican 229 (6). 110-117. Goldman-Eisler, F. 1968. Psycholinguistics: Experiments in  Spontaneous Speech. London: Academic Press. Grinker, R.R., and P.C. Bucy. 1949. Neurology. Springfield, 111.: Charles C Thomas. Haggard, M. 1973. Abbreviation of consonants in English pre-and post-vocalic clusters. Journal of Phonetics 1, 9-24. 82 Kardy, J.C. 1970. Development of neuromuscular systems underlying speech production. Speech.and the Dentofacial  Complex: The State of the Art (ASHA Reports No. 5), 49-68. Washington, D.C.: American Speech and Hearing Association. Johnston, T.B., and J. Whillis (Eds.). 1954. Gray's Anatomy  Descriptive and Applied, 31st edition. London: Longmans, Green and Co. Kent, R.D., and K.L. Moll. 1975. Articulatory timing in selected consonant sequences. Brain and Language 2. 304-323. Konigsmark, B.'W. 1970. Neuroanatomy of speech. Speech and  the Dentofacial Complex: The State of the Art (ASHA Re-ports No. 5), 3-19. Washington, D.C.: American Speech and Hearing Association. Kozhevnikov, V.A., and L.A. Chistovich. 1965. Speech: Articulation and Perception. [Joint Publications Research Service, No. 30,543,] Washington, D.C.: U.S. Department of Commerce. Lee, B.S. 1950. Effects of delayed speech feedback. Journal  of the Acoustical Society of America 22. 824-826. Lee, B.S. 1951. A r t i f i c i a l stutter. Journal of Speech and  Hearing Disorders 16. 53-55. Lehiste, I. 1971. Temporal organization of spoken language. Form & Substance, ed. by L.L. Hammerich, R. Jakobson, and E. Zwirner, 159-169. Copenhagen: Akademisk Forlag. Lehiste, I., and G.E. Peterson. 1959. Vowel amplitude and phonemic stress in American English. Journal of the  Acoustical Society of America 31. 428-435. Lenneberg, E.H. 1967. Biological Foundations of Language. New York: Wiley. MacKay, D.G. 1970. Spoonerisms: The structure of error in the s e r i a l ordering of speech. Neuropsychologia 8. 323-350. 83 MactJeilage, P.F. 1970. Motor c o n t r o l of s e r i a l o r d e r i n g of ". •speech. P s y c h o l o g i c a l Review 77. 182-196. MacNeilage, P.F. 1972. Speech physiology- Speech and C o r t i c a l F u n c t i o n i n g , ed. by J . G i l b e r t , 1-72. New York: Academic Press. Meringer, K. 1908. Aus dem Leben der Sprache. B e r l i n . [ C i t e d i n Fromkin, 1971.] Meringer, K., and K. Mayer. 1895. Versprechen und Verlesen, Eine p s y c h o l o g i s c h l i n g u i s t i s c h e Studie.. Vienna. [ C i t e d i n Fromkin, 1971.] Merton, P.A. 1953. Speculations on the s e r v o - c o n t r o l of movement. The S p i n a l Cord, ed. by G.E.W. Wolstenholme, 247-255. London: C h u r c h i l l . [ C i t e d i n Abbs, 1973.] Mortimer, E.M., and K. Akert. 1961. C o r t i c a l c o n t r o l and r e p r e s e n t a t i o n of f u s i f o r m neurons. American J o u r n a l  of P h y s i c a l Medicine 48. 228-248. [ C i t e d i n Abbs, 1973.] Ne t t e r , F.H. 1972. The CIBA C o l l e c t i o n of Medical I l l u s t r a -t i o n s , Volume 1: Nervous System. Summit, N.J.: CIBA Pharmaceutical Co. Nooteboom, S.G. 1969. The tongue s l i p s i n t o p a t t e r n s . Nomen: Leyden Studies i n L i n g u i s t i c s and Phonetics, ed. by A.G. Sciarone et a l . , 114-132. The Hague: Mouton. Ohala, J . J . 1970. Aspects of the c o n t r o l and production of speech. Working Papers i n Phonetics, No. 15. Los Angeles: U n i v e r s i t y of C a l i f o r n i a . Ohala, J . J . 1973. The temporal r e g u l a t i o n of speech. C o l l e c t i o n of Papers Presented at Symposium on A u d i t o r y  A n a l y s i s and P e r c e p t i o n of Speech (Leningrad, August 21-24), ed. by H. F u j i s a k i , 516-538. Tokyo: U n i v e r s i t y of Tokyo. P e n f i e l d , W., and T. Rasmussen. 1950. The C e r e b r a l Cortex of Man. New York: Macmillan. 84 Perifield, W., and L. Roberts. 1959. Speech and Brain-Mechanisms . Princeton, N.J.: Princeton University Press. Ruch, T.C., H.D. Patton, J.W. Woodbury, and A.L. Towe. 1967. Neurophysiology. Philadelphia: W.B. Saunders. [Cited in Abbs, 1973.] Sataloff, J. 1966. Hearing Loss. Philadelphia: Lippincott. Schwartz, M.F. 1970. Duration of /s/ in /s/-plosive blends. • Journal of the Acoustical Society of America 47. 1143-1144. Sears, T.A., and J. Newsome Davis. 1968. The control of respiratory muscles during voluntas breathing. Sound  Production in Man, ed. by A. Bouhuys, 183-190. New York: New York Academy of Sciences. [Cited in Abbs, 1973.] Smith, J.L. 1969. Fusiform Neuron Block and Voluntary Arm  Movement in Man. Unpublished doctoral dissertation, University of Wisconsin. [Cited in Abbs, 1973.] Smith, T.S. 1973. Review of Bowman (1971). Journal of  Phonetics 1. 171-179. Stark, L. 1968. Neurological control systems: Studies in bioengineering. The Hand, 301-410. New York: Plenum. [Cited in Abbs, 1973.] Stromsta, C. 1962. Delays associated with certain sidetone pathways. Journal of the Acoustical Society of America 34. 392-396. [Cited in Van Riper, 1971.] Sturtevant, E.H. 1917. Linguistic Change. Chicago: Univer-sity of Chicago Press. Sturtevant, E.H. 1947. An Introduction to Linguistic Science. New Haven: Yale University Press. Van Riper, C. 1971.. The Nature of Stuttering. Englewood C l i f f s , N.J.: Prentice-Hall. APPENDIX A The f i r s t paragraph of nThe Rainbow Passage'' (Fairbanks, 1960, p. 127): "When the s u n l i g h t s t r i k e s raindrops i n the a i r , they act l i k e a prism and form a rainbow. The rainbow i s a d i v i s i o n of white l i g h t i n t o many b e a u t i f u l c o l o r s . These take the shape of a long round arch, w i t h i t s path high above, and i t s two ends apparently beyond the h o r i z o n . There i s , according to legend, a b o i l i n g pot of gold at one end. People look, but no one ever f i n d s i t . When a man looks f o r something beyond h i s reach, h i s f r i e n d s say he i s l o o k i n g f o r the pot of go l d at the end of the rainbow." 85 APPENDIX B Paragraphs w i t h embedded w o r d - i n i t i a l / s C / - c l u s t e r s : /sp-/: " I n school I had a f r i e n d nicknamed 'Spud', who was t e r r i b l e at most everything. One afternoon i n the school s p e l l i n g bee, Spud s p e l t ' s p a g h e t t i ' s p e e d i l y and won the school s p e l l i n g p r i z e . " / s t - / : " I n school I a l s o had two f r i e n d s , Stan and Stewart, who were always f i g h t i n g w i t h one another. One day w h i l e i n a f i g h t , Stewart stepped on Stan's  stomach, and he d i e d three days l a t e r . " /sk-/: "Another f r i e n d from my school days was Skana. She l i k e d doing anything b e t t e r than going to school. In f a c t , Skana skipped school so s k i l l f u l l y that no one knew what happened to her." 86 APPENDIX C "Tongue-twisters" w i t h embedded w o r d - i n i t i a l / s C / - c l u s t e r s /sp-/: "A spectre of a s p i r i t e d s p e c t a c l e d Spanish 'Spartan' c a l l e d Spinoza ate the spotted spiced spinach s p o r a d i c a l l y w i t h a spoon before i t s p o i l e d . " / s t - / : "The s t a l w a r t s t a l l i o n and the statuesque.,£teer were s t a r t l e d by the s t o i c a l s t a b l e s t a f f s t e a l i n g sjtagnant stew o f f the stove." /sk-/: "Scandinavians s k i l l f u l l y s k i n and s c a l d s c a l l i o n s , s c a l l o p s , scampies and skimpy scorpions u n t i l they are s c a r c e l y s c a r l e t , then s c a r f them down." 87 APPENDIX D PHONETIC TRANSCRIPTIONS.AND"CODING The repetition errors for each subject in this study are l i s t e d according to type (or experimental group), phonetic transcription (using a modified version of the International Phonetic Alphabet), and standard orthographi form of the target utterance. A l l errors involve the word i n i t i a l clusters: /sp-/, /st-/, /sk-/. Page Svibject 1 8 9 Subject 2 90 Subject 3 91 Subject 4 .. 92 Subject 5 93 Subject 6 94 88 SUBJECT 1. REPETITION ERRORS. (Total: 18) Experimental Group #1: (Phoneme Repetition) Experimental Group #2: (Cluster Repetition, with a Pause) Experimental Group #3: (Cluster Repetition, without a Pause) Experimental Group #4: (/sCV-/ Syllable Repetition) Experimental Group #5: (/sCVC-/ Syllable Repetition) Experimental Group #6: (Word Repetition) /sp-/ [s?sp inooza] 'Spinoza' /st-/ [ s ? s t o o i k J ] 'stoical' /sk-/ [sk?sks3mpiz] 'scampies' [skh?ski I f o li] ' s k i l l f u l l y ' [ skh?skcsmp i z ] 'scampies' [ s k ? s k i I f o l i ] ' s k i l l f u l l y ' /sp-/ [ s p s p o j a d i k l i ] 'sporadically' /st-/ [ s t h s t f f i tJ juesk] 'statuesque' /sk-/ [ s k s k i I f o l i ] ' s k i l l f u l l y ' /sp-/ [spe*?spadid] 'spotted' [ s p e ? s p e k t h a J ] 'spectre' [sphei?spadid ] 'spotted' /st-/ [ s t h e ? s t a j t I d ] 'startled' /sk-/ [ skha? ska j l e t ] 'scarlet' /st-/ [st h aeks?st*t J juesk] 'statuesque' [ stffit?staet J juesk] 'statuesque' /sk-/ [ skaj?skaj 1 et ] 'scarlet' [skaem?skajf ] 'scarf' No observations. 89 SUBJECT 2. REPETITION ERRORS . (Total: 28) Experimental Group #1: /.sp-/ [s?spekt3kald] 'spectacled' (Phoneme Repetition) /st-/ [ s?siaet J juesk] 'statuesque' /sk-/ [ s?sks i jenz ] 'scallions' [s?skimp i ] 'skimpy' [ s ? s k h a j l e t ] 1 scarlet' [ s?skae 1 aps ] 'scallops' [s?skimp i ] 'skimpy' [ s?skae 1 janz ] 'scallions' [ s?skaemp i z ] 'scampies' Experimental Group #2: /sp-/ [s *ph?spoj£edi kl i'] 1 sporadically' (Cluster Repetition., /sk-/ [ s k h ? s k iIfoli] ' s k i l l f u l l y ' with a Pause) [sk?sksl j anz ] 'scallions' Experimental Group #3: /st-/ C S t S t h1 1t Q ] 'stealing' (Cluster Repetition, /sk-/ [ sk: skin ] 'skin' without a Pause) Experimental Group #4: /sp-/ [ s p h a i ? s p a d i d ] 'spotted' (/sCV-/ Syllable /st-/ [ st hse?stst J juesk] 'statuesque' Repetition) [stestffi1jan] 'stallion' [st ho ? s t o o t k l ] 'stoical' /sk-/ [ s t i ? s k i I f o l i ] ' s k i l l f u l l y ' Experimental Group #5: /sp-/ [sph i j?sp^ i J i d i d ] 'spirited' (/sCVC-/ Syllable [ s p e k h ? s p e k t a J ] 'spectre' Repetition) /st-/ [ s t h o o i k s t h o o i k h J ] 'stoical' /sk-/ [ s t i l f ? s k i I f o l i ] ' s k i l l f u l l y ' [ skirnp?skimp i ] 'skimpy' [skasmph?skaemp i z ] 'scampies' [skaenth?sk3empiz] 'scampies' [ s k a j ? s k e u s l i ] 'scarcely' Experimental Group #6: /sp-/ [ spekld?spekt haBkl d ] 'spectacled' (Word Repetition) 90 SUBJECT 3. REPETITION Experimental Group #1: (Phoneme Repetition) Experimental Group #2: (Cluster Repetition, with a Pause) Experimental Group #3: (Cluster Repetition, without a Pause) Experimental Group #4: (/sCV-/ Syllable Repetition) Experimental Group #5: (/sCVC-/ Syllable. Repetition) Experimental Group #6: (Word Repetition) ERRORS. /sk.-/ [ s?skimph i ] [s?skimp i ] . [s?skasleps] [s?skhimp i ] [ s ? skaj f ] /sk-/ [ skh?ska3ndenetv i janz ] [skh?skbaemp i z ] /sk-/ [ s k r h s k a j f ] [ s k : sk i I f o l i ] /sk-/ [ sklu?sk h imp i ] [skha?skha-i let] / s t - / '[ staeg?staet J juesk] / sk- / [ s ko jp i z ? sk s l ap s ] [ s kh in ? ska j f ] [ skh in?skh in] (Total: 15) ' skimpy' ' skimpy' 'scallops' ' skimpy' 1 s c a r f ' Scandinavians' 'scampies' 'scarf ' s k i l l f u l l y ' ' skimpy' 'scarlet' ' statuesque' ' scallops' 'scarf 'skin' 91 SUBJECT 4. REPET3 Experimental Group #1: (Phoneme Repetition) Experimental Group #2; (Cluster Repetition, with a Pause) Experimental Group #3: (Cluster Repetition, without a Pause) Experimental Group #4: (/sCV-/ Syllable Repetition) Experimental Group #5: (/sCVC-/ Syllable Repetition) Experimental Group #6: (Word Repetition) ERRORS. (Total: 33) /sp-/ [ ; s ? spektskMd] 'spectacled' 1st-! i ; s ? s teb l ] 'stable' ! s?stae 1 jan ] 'stallion' /st-/ [ ! s t h ? s t e i b l 3 'stable' ! s t h ? s t a l w e J t ] 'stalwart' . s t h ? s t o o i k h a l ] 'stoical' .st h?stffignt ] 'stagnant: ; s t h ? s t h f l i Q ] 'stealing' . s k h ? s k m ] 'skin' .sk: h skimp i ] 1 skimpy' .sk?skojp i j e n z ] 'scorpions' .sk:?sktmpi] 'skimpy' . skh?skasl jenz ] 'scallions' /sp-/ [ . sp:sphojsd ik l i ] 'sporadically' . sp : h spad id ] 'spotted' .sp:spaj?n ] 'Spartan' /sk-/ [ . s k : h s k i I f o l i ] ' s k i l l f u l l y ' /sp-/ [ . spaTspasm J ] 'Spanish' /st-/ [ stae?sthae 1 jen ] 'stallion' . s t e ? s te ib ] ] 'stable' st3B?staet Jjuesk'] 'statuesque' s t i ? s t oov ] 'stove' stae?staegnt ] 'stagnant' sta?sta IweJt] 'stalwart' /sk-/ [ skeA?ska31 jenz ] 'scallions' /sp-/ [ spad ? sp i j i d i d ] 'spirited' /st-/ [ staegn?sta3t J juesk] 'statuesque' s t a ljeJ ? s t a l w e J t ] 'stalwart' /sk-/ [ ski 1?skimpi] 'skimpy' s k i l ? s k i n ] •skin' /sk-/ [ s kha j f e t ? s kha j l e t ] 'scarlet 1 skhuiph9?skimp i ] 'skimpy' ski 1f?skimpi] 'skimpy' 92 SUBJECT 5. REPETITION ERRORS. (Total: 11) Experimental Group #1: • Experimental Group #2: Experimental Group #3: (Cluster Repetition, without a Pause) Experimental Group #4: (/sCV-/ Syllable Repetition) Experimental Group #5: (/sC«'C-/ Syllable Repetition) Experimental Group #6: (Word Repetition) No observations. No observations. /sk-/ [sk:ska laps] 'scallops' /sp-/ [ spA ? spad id] 'spotted' /st-/ [ s t h o o ? s t h a 31 Jj'uesk] 'statuesque' /sk-/ [skhae?skae laps] 'scallops' [skae?skas!eps] 'scallops 1 [sk*u?skhin] 'skin' /sk-/ [ skhaem?ska 1 aps J 'scallops' [sksel?sk3ndmevi janz ] 'Scandinavians' [ s t a j ? s k h a j l e t ] 'scarlet' /sp-/ [ s p e k t h a ? s p e k t h a J ] 'spectre' /sk-/ [ s t h e j s l i s k h e u s l i ] 'scarcely' 93 SUBJECT 6. REPETITION ERRORS. (Total: 13) Experimental Group #1 No observations. Experimental Group #2: /sp-/ [ spVsp i - H d i d ] (Cluster Repetition, [ s p h ? s p e k t h a k aId ] with a Pause) r ho n ' |. sp h?spaemj J [sph?spajtn] /sk-/ [skh?5kaleps] [sk^?sk in] [ s k h ? s k i I f o l i ] Experimental Group #3: /st-/ [ s t : staetJjuesk] (Cluster Repetition, without a Pause) Experimental Group #4: /st-/ [ s te?sta I we-11 ] (/sCV-/ Syllable /sk-/ [ skh i ? sk impi ] Repetition) [ska?skaIeps] [skas?skeel j enz ] Experimental Group #5: /st-/ [ staeg?staegnant ] C/sCVC-/. Syllable Repetition) Experimental Group #6: No observations. 'spirited' 'spectacled' 'Spanish' 'Spartan' ' scallops' •skin' ' s k i l l f u l l y ' ' statuesque' 'stalwart' 'skimpy' 'scallops' 'scallions' 'stagnant' 94 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0093697/manifest

Comment

Related Items