Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of coarticulation on the role of transitions in vowel perception Ukrainetz, Teresa A. 1987

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

Item Metadata

Download

Media
831-UBC_1987_A6_7 U37.pdf [ 4.6MB ]
Metadata
JSON: 831-1.0097195.json
JSON-LD: 831-1.0097195-ld.json
RDF/XML (Pretty): 831-1.0097195-rdf.xml
RDF/JSON: 831-1.0097195-rdf.json
Turtle: 831-1.0097195-turtle.txt
N-Triples: 831-1.0097195-rdf-ntriples.txt
Original Record: 831-1.0097195-source.json
Full Text
831-1.0097195-fulltext.txt
Citation
831-1.0097195.ris

Full Text

THE EFFECT OF COARTICULATION ON THE ROLE OF TRANSITIONS IN VOWEL PERCEPTION By TERESA A. UKRAINETZ B.ft., U n i v e r s i t y o f C a l g a r y , 1984 ft THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES S c h o o l o f ftudiology and Speech S c i e n c e s We a c c e p t t h i s t h e s i s a s c o n f o r m i n g t o t h e r e q u i r e d s t a n d a r d UNIVERSITY OF BRITISH COLUMBIA J u l y , 1987 © T e r e s a A. U k r a i n e t z , 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6(3/81) ABSTRACT The present study examines the e f f e c t of context on the use of t r a n s i t i o n s as cues t o vowel p e r c e p t i o n . T h i r t y V j C V a C V I u t t e r a n c e s were recorded, with V» being one of the thr e e vowels / a , i , u / , and V a one of ten E n g l i s h vowels (/ i , I, e l , E, ae, a, "", oU, U, u/) . A f t e r removal of the outer vowels < V t ) , t h r e e s e t s o f s t i m u l i were c r e a t e d from the C V a C p a r t s : ( 1 ) unmodified c o n t r o l s ( C O ) 5 ( £ ) V a s t e a d y - s t a t e only (SS); and ( 3 ) t r a n s i t i o n s only (TR)• Twenty s u b j e c t s were asked t o i d e n t i f y V s . S u b j e c t s and speaker were matched f o r d i a l e c t and a l l s u b j e c t s had some phonetics t r a i n i n g . R e s u l t s showed s i g n i f i c a n t d i f f e r e n c e s a c r o s s c o n d i t i o n s and contexts. Scores f o r SS s t i m u l i , f o r a l l contexts, were as high as f o r CO s t i m u l i . Performance on the TR s t i m u l i was as good as on the other two c o n d i t i o n s f o r two of the contexts. However, f o r the TR c o n d i t i o n — / a / context, performance was c o n s i d e r a b l y worse than f o r any other combination of c o n d i t i o n s and contexts. P o s s i b l e reasons f o r t h i s are di s c u s s e d , and the need f o r t e s t i n g of other vowel c o n t e x t s i s emphasised. It i s concluded that, i n some V i C V a C V t contexts, t r a n s i t i o n s can provide i n f o r m a t i o n about vowel i d e n t i t y on a l e v e l equal t o s t e a d y - s t a t e alone, or t o the combined i n f o r m a t i o n provided by both t r a n s i t i o n s and s t e a d y - s t a t e s . T h i s e f f e c t , however, i s not uniform a c r o s s contexts. For at l e a s t one context, t r a n s i t i o n s alone are not s u f f i c i e n t t o cue vowel i d e n t i t y at a l e v e l comparable to s t e a d y - s t a t e or combined i n f o r m a t i o n . T h i s lack of u n i f o r m i t y suggests that the r o l e of t r a n s i t i o n s v a r i e s with the type of vowel context present, and c o n c l u s i o n s about general u s e f u l n e s s await s y s t e m a t i c t e s t i n g of a number of vowel contexts. i v TABLE OF CONTENTS Page ABSTRACT, . i i TABLE OF CONTENTS. . . . . i v LIST OF TABLES...............................................vi ACKNOWLEDGEMENT............................................viii 1„ INTRODUCTION. .............................................. 1 £. REVIEW OF THE LITERATURE. ................................... 3 £„ 1 I n t r o d u c t ion. ............................................ 3 £. £ Cues t o vowel percept ion. .............................3 £.3 C o a r t i c u l a t ion. ...................................... 15 £.4 P e r c e p t u a l e f f e c t s of c o a r t i c u l a t i o n . £1 3. AIMS OF THE EXPERIMENT. .................................... £6 4. METHODOLOGY...............................................£8 4. 1 Preparat ion of the t e s t tapes. ........................ £8 4. £ Experimental cond i t ions. ............................... 38 4.3 Subje c t s . ............................................ 39 4. 4 Procedure. ............................................. 40 5. RESULTS................................................... 41 5. 1 Organizat ion of the data. ............................. 41 5. £ Performance over cond i t ions and contexts. ............. 44 5.3 Comparison of medial / i / and /u/ groupings........... 51 5.4 Item A n a l y s i s . .52 6. DISCUSSION. 55 6. 1 Vowel cues avai l a b l e i n TR st irnu.1 i . . . . . . . . . . . . . . . . . . . . 55 6.3 The e f f e c t of cut point placement... 6.4 Poor performance i n the /a/ context 6. 5 Summary and c o n c l u s i o n s . . . . . . . . . . . . REFERENCES CITED........................... APPENDIX.................................. . v i LIST OF TABLES Table Page I. Range of vowel and vowel component d u r a t i o n s ( i n msec.) f o r the present data and f o r the Strange et a l . (1983) data. ............................. 31 II . R e l a t i v e d u r a t i o n ( i n "/ of t o t a l vowel d urat i on) of i n i t i a l and f i na1 t r a r e i t i ons obtained with two methods. 34 I I I . Re 1 at i ve d ur at ions ( i n "/. o f t ot a 1 vowe 1 d u r a t i o n ) of i n i t i a l and f i n a l t r a n i s i t ions chosen f o r each vowel 36 IV. Vowel i n t e n s i t i e s ( i n dB) a f t e r adjustment ac c o r d i n g t o Peterson and L e h i s t e (1959)............... 38 V. Summated c o n f u s i o n matrices f o r a l l su b j e c t VI. E r r o r r a t e ( i n °A) a c r o s s c o n d i t i o n s and contexts. 45 VII. M o d i f i e d e r r o r r a t e s ( i n "/•) a c r o s s c o n d i t ions and c o n t e x t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 V I I I . ANOVA r e s u l t s f o r both unmodified and modified t o t a l s . 49 IX. Comparison of e r r o r r a t e s ( i n "/.) f o r two groupins of / i / and /u/............................5£ X. E r r o r r a t e s ( i n %) as a f u n c t i o n of vowel type f o r a l l co n d i t ions and c o n t e x t s . . . . . . . . . . . . . 53 X I. Compar i son of ident i f i cat i on responses Vi CVaCVi t r a n s c r i p t i o n and - C V e C - f o r c e d i c e s f o r the problem vowels......................... 60 v i i i ACKNOWLEDGEMENT I w o u l d l i k e t o e x p r e s s my a p p r e c i a t i o n t o a l l t h o s e p e o p l e who p r o v i d e d me h e l p and e n c o u r a g e m e n t t h r o u g h o u t t h e w r i t i n g o f t h i s t h e s i s : - my s u p e r v i s o r , Dr. A n d r e - P i e r r e B e n g u e r e l 5 - my c l a s s m a t e s and s u b j e c t s (who were k i n d l y one i n t h e s a m e ) ; - C a r o l and L i s a , w i t h o u t whom, t h i s t h e s i s w o u l d h a v e been a rnuch 1 one 1 i e r e n d e a v o r 5 - and D a v i d , whose l o v e and s u p p o r t , a s w e l l a s u n f l a g g i n g i n t e r e s t i n my work, h e l p e d k e e p me g o i n g t h e s e two y e a r s . 1 Chapter 1 INTRODUCTION Speech p e r c e p t i o n has o f t e n been examined by use of short, i s o l a t e d phoneme or s y l l a b l e - l e n g t h segments, n a t u r a l l y or s y n t h e t i c a l l y produced- G e n e r a l i z a t i o n s were then made to our comprehension of speech sounds i n connected d i s c o u r s e . T h i s approach has tended t o assume that speech p e r c e p t i o n i s a s e q u e n t i a l process, where the l i n k i n g of d i s c r e t e phonemes or s y l l a b l e s i n words and sentences has no e f f e c t on the nature of those b a s i c speech u n i t s , and that the p e r c e p t i o n of each unit i s e s s e n t i a l l y u n a f f e c t e d by the preceding or f o l l o w i n g segments. Descr i pt i ons of percept ion and phoneme i dent i f i c a t i on based so 1 e 1 y on i so 1 at ed prod uct i ons may be rn i s 1 ead ing s i nee evidence from s t u d i e s i n speech p r o d u c t i o n suggest that the b a s i c speech u n i t s are produced i n a temporally o v e r l a p p i n g manner. T h i s o v e r l a p , or c o a r t i c u l a t i o n , i s due to planning and i n e r t i a l f a c t o r s , r e s u l t i n g i n i n d i v i d u a l phonemes whose phonetic r e a l i s a t i o n s are q u i t e d i f f e r e n t i n s p e c t r a l and temporal c h a r a c t e r from when they occur i n i s o l a t i o n . The c o a r t i c u1at i on t a k i n g p1ace i n the prod uct i on process i s r e f l e c t e d i n the r e s u l t i n g a c o u s t i c p a t t e r n . In some cases, the e f f e c t o f c o a r t i c u l a t i o n can be seen more than a s y l l a b l e away on sound spectrograms. I f the changes brought about by c o a r t i c u l a t i o n are v i s i b l e and measurable, i t i s p l a u s i b l e that they may be used by the perc e p t u a l system and they could be reasonably expected t o a f f e c t the nature of p e r c e p t i o n . Evidence suggests that, i n p e r c e i v i n g speech, we do seem to make use of many of the cues a v a i l a b l e , even those r e l a t i v e l y d i s t a n t l y l o c a t e d . I d e n t i f i c a t i o n of speech segments i n v o l v e s a dynamic i n c o r p o r a t i o n of i n f o r m a t i o n over a pe r i o d of time t o a r r i v e at a u n i f i e d percept of a /b/ or an /a/. The " d i s t o r t i o n " o f the phoneme caused by t h i s temporal and s p e c t r a l spread of cues would be expected t o a f f e c t which cues are used and t h e i r r e l a t i v e importance f o r phoneme i d e n t i f i c a t i o n as compared t o the o r d e r i n g o f cues e s t a b l i s h e d f o r phone or s y l l a b l e — l e n g t h segments. Pl q u e s t i o n a r i s e s as t o how the changes brought about by c o a r t i c u l a t i o n a f f e c t the cues normally a v a i l a b l e when the phone or s y l l a b l e i s presented i n i s o l a t ion. The present study looks at the cues used i n the p e r c e p t i o n of vowels, and how the presence of context, through the r e s u l t a n t c o a r t i c u l a t i o n , a f f e c t s the r e l a t i v e u t i l i t y of t r a n s i t i o n s and s t e a d y — s t a t e i n f o r m a t i o n as cues t o vowel ident i t y . C h a p t e r £ REVIEW O F THE LITERATURE S. 1 I n t r o d u c t i on Modern v o w e l p e r c e p t i o n s t u d i e s d a t e f r o m t h e i n v e n t i o n o f t h e s p e c t r o g r a p h n e a r l y f o r t y y e a r s ago. A l t h o u g h t h e number o f s t u d i e s done s i n c e t h e n i s v e r y l a r g e , o u r u n d e r s t a n d i n g o f v o w e l p e r c e p t i o n i s s t i l l l i m i t e d . S e c t i o n £. £ s u m m a r i z e s t h e s t a t e o f k n o w l e d g e o f v o w e l p e r c e p t i o n and t h e e f f e c t s o f c o n t e x t on p e r c e p t i o n . S e c t i o n £.3 f u r t h e r e x a m i n e s t h e i n f l u e n c e o f c o n t e x t and i t s r e s u l t a n t c o a r t i c u l a t i o n , a t t h e a r t i c u l a t o r y and a c o u s t i c l e v e l s . S e c t i o n £.4 d e a l s w i t h t h e p e r c e p t i o n o f c o a r t i c u l a t i o n , and s u g g e s t s how t h e p e r c e p t u a l e f f e c t o f c o a r t i c u l a t i o n on s p e e c h s e g m e n t s may a f f e c t o u r c u r r e n t c o n c e p t i o n o f t h e r e l a t i v e u t i l i t y o f known v o w e l c u e s . £. £ C u e s t o Vowel P e r c e p t i o n V o w e l s h a v e been t r a d i t i o n a l l y c h a r a c t e r i z e d by t h e f r e q u e n c i e s o f t h e i r f i r s t few f o r r n a n t s , e s p e c i a l l y t h e f i r s t t wo i n t h e c a s e o f E n g l i s h ( r e f e r r e d t o a s F t and F s>» When v o w e l s a r e p r o d u c e d i n i s o l a t i o n , t h e f o r r n a n t f r e q u e n c y v a l u e s o f t h e s e v o w e l s , f o r a s p e a k e r , c a n be p l o t t e d a s p o i n t s i n a t w o - d i m e n s i o n a l s p a c e ( F t v s . F e ) . Forrnant p l o t t i n g t e n d s t o c l u s t e r p h o n e m i c a l l y e q u i v a l e n t t o k e n s , and c a n be r e l a t e d t o a r t i c u l a t o r y p r o d u c t i o n p a t t e r n s . T h i s c h a r a c t e r i z a t i o n , h o w e v e r , i s i n s u f f i c i e n t t o d i s t i n g u i s h v o w e l s i n a l l but t h e most i d e a l c o n d i t i o n , that of a s i n g l e speaker producing i s o l a t e d vowels. Peterson and Barney (1352) p l o t t e d the vowels of IB /hVd/ words spoken by 76 speakers, i n an F t ~ - F a graph. They found that the d i s t r i b u t i o n s o f forrnant v a l u e s d i d not correspond c l o s e l y t o the d i s t r i b u t i o n of phonemic values, r e s u l t i n g i n o v e r l a p p i n g vowel loops. Stevens and House (1963) measured the f i r s t two forrnant f r e q u e n c i e s and bandwidths of e i g h t vowels, f o r t h r e e male American E n g l i s h speakers, i n #V#, /hVd/, and 14 d i f f e r e n t /hSCVC/ c o n t e x t s by t h r e e male speakers. They found s i g n i f i c a n t d i f f e r e n c e s between t h e i r r e s u l t s and the vowel loops p l o t t e d by Peterson and Barney (1952), and by Peterson (1961) f o r vowels i n /hVd/ context and i n i s o l a t i o n r e s p e c t i v e l y . Some of the d i f f e r e n c e was s y s t e m a t i c and found t o be caused by d i f f e r e n c e s i n p r o d u c t i o n r e l a t e d t o v o c a l t r a c t length, e s p e c i a l l y f o r the vowels i n i s o l a t i o n and i n /hVd/ context. Consonants were found t o exert a l a r g e e f f e c t on vowel forrnant values. The consonant e f f e c t s were s y s t e m a t i c and r e g u l a r , depending on plac e of a r t i c u l a t i o n , manner of a r t i c u l a t i o n and presence of v o i c i n g as w e l l as on the p a r t i c u l a r vowel i n v o l v e d . The a c o u s t i c e f f e c t s tended to be of movement towards the locus of the consonant or movement towards n e u t r a l schwa values. Movement towards the c e n t e r of the vowel t r i a n g l e , or c e n t r a l i z a t i o n , has a l s o been found t o occur with i n c r e a s e d speech r a t e s and i n vowels of weakly 5 s t r e s s e d s y l l a b l e s , both i n B r i t i s h and American E n g l i s h (Lindblom, 19633 Gay, 1978). I n d i v i d u a l d i f f e r e n c e s i n degree of F a undershoot not accounted f o r by s y l l a b i c r a t e or v o c a l t r a c t d i f f e r e n c e s have a l s o been documented (Stevens, House & Paul, 1966), but, i n general, the d i f f e r e n c e s i n forrnant frequency o c c u r r i n g f o r the same vowel generated by d i f f e r e n t speakers can be a s c r i b e d to d i f f e r e n c e s i n v o c a l t r a c t s i z e , speaking r a t e and degree of s t r e s s . Stevens and House (1963) concluded that the changes i n measured forrnant frequency v a l u e s f o r a given vowel are ... i n l a r g e measure e x p l i c a b l e i n terms of the dynamic p r o p e r t i e s of the a r t i c u l a t o r y mechanism. The i n e r t i a and delay c h a r a c t e r i s t i c s of the a r t i c u l a t o r y s t r u c t u r e s and musculature r e s u l t i n undershoot i n the motion of the s t r u c t u r e s from one t a r g e t c o n f i g u r a t i o n to the next, g i v i n g r i s e t o corresponding changes i n the v o c a l t r a c t resonances. (p. 126) Although changes i n vowel forrnant frequency l a r g e l y f o l l o w l a w f u l and p r e d i c t a b l e p a t t e r n s based on a r t i c u l a t o r y dynamics, the problem of how they are decoded by the l i s t e n e r remains. C h a r a c t e r i z a t i o n s based on s i n g l e forrnant frequency measures (e.g. F e values) r e s u l t i n too great an o v e r l a p i n c a t e g o r i e s to adequately account f o r vowel i d e n t i f i c a t i o n . As a r e s u l t of t h i s ambiguity, a t t e n t i o n was devoted to f u r t h e r s p e c i f y i n g the s t a t i c model of vowel space as well as to i n v e s t i g a t i n g other p o s s i b l e cues w i t h i n the vowel s i g n a l . An e a r l y attempt to r e f i n e the vowel space model was presented by Peterson (1961). Peterson recorded s u s t a i n e d vowels i m i t a t e d by men, women and c h i l d r e n . The speakers l i s t e n e d to recorded r e f e r e n c e vowels and attempted t o produce a phonetic match. The vowels spoken i n c l u d e d non-English vowels, which Peterson r e p o r t e d the s u b j e c t s had no d i f f i c u l t y matching. Pi s i n g l e frequency measure was taken f o r each of the f i r s t two formants. V a r i o u s t r a n s f o r m a t i o n s using l i n e a r , l o g a r i t h m i c , and rnel s c a l e s were attempted, so as to produce groupings of phonet i ca11y equi vaIent t okens and d i f f e r e n t i a t ion of non-e q u i v a l e n t items. Peterson (1961) concluded that, f o r any of these t r a n s f o r m a t i o n s , " n e i t h e r a f i x e d forrnant frequency nor a forrnant frequency r a t i o h y p o t h e s i s i s adequate t o e x p l a i n vowel p e r c e p t i o n f u l l y " ( i b i d , p. 24-S5). He a l s o concluded that o t h e r a s p e c t s of the vowel i n f l u e n c e p e r c e p t i o n , such as the fundamental frequency, amplitude and phonetic environment. Stevens and House (1963) suggested that some of the d i f f i c u l t y s p e c i f y i n g a workable model a r i s e s from the lack of agreed upon method f o r sampling vowels and the i n s u f f i c i e n c y of" using a s i n g l e f r e q uency meas ur e t o ch ar act e r i z e forrnant s wh i ch sh ow continuous movement throughout the s y l l a b l e nucleus. Lindblom (1963) advocated d e f i n i n g the t a r g e t not as the measured forrnant frequency value, but as a p o t e n t i a l l y u n r e a l i z e d asymptotic value. T h i s value corresponds t o the average i s o l a t e d vowel p r o d u c t i o n value f o r that speaker. Stevens and Houses (1963), as mentioned e a r l i e r i n t h i s s e c t i o n , confirmed the s y s t e m a t i c and p r e d i c t a b l e nature of the changes i n vowel forrnant f r e q u e n c i e s . Transformations o f the t a r g e t forrnant f r e q u e n c i e s and the r e l a t i v e r e l a t i o n s h i p s between them were developed t o account f o r d i f f e r e n c e s among speakers (e.g. Neary, 1977; Skinner, 1977). Approaches i n v o l v i n g c a l i b r a t i o n s of a p a r t i c u l a r speaker's vowel space e i t h e r from f r e q u e n c i e s of the forrnants of other vowels o c c u r r i n g i n the same a u d i t o r y context, or from high forrnants (F* or F 3 ) have been p o s t u l a t e d , e.g. by Joos (1948). Joos, as r e p o r t e d i n Ladefoged and Broadbent (1957), suggested that vowel i d e n t i f i c a t i o n was based on knowledge of that speaker's vowel space. He suggested f u r t h e r that a vowel was ident i f ied, not on the b a s i s of i t s abso1ute values, but on the r e l a t i o n s between those vowel forrnant f r e q u e n c i e s and the general ranges of f r e q u e n c i e s which seemed t y p i c a l f o r that speaker. Ladefoged and Broadbent (1957) supported t h i s c l a i m by showing, with s y n t h e t i c speech, that i d e n t i f i c a t i o n of a t e s t word i s g r e a t l y i n f l u e n c e d by the range of forrnant f r e q u e n c i e s i n the preceding c a r r i e r phrase. Gerstman (1968) was a b l e t o s u c c e s s f u l l y develp a computer a l g o r i t h m using the extreme v a l u e s of a speaker's vowel space ( u s u a l l y the vowels / a / , / i / , / u / ) to s c a l e a l l the t a l k e r ' s vowels. Whether or not t h i s type of s c a l i n g i s an a c t u a l p e r c e p t u a l s t r a t e g y i s as yet unknown, but other s t u d i e s , such as Vertarugge, Strange, Shankweiler and Edrnan (1976), Macchi (1980), and Assrnan, Neary, and Hogan (198S) have shown th a t , although speaker f a m i l i a r i t y does improve i d e n t i f i c a t i o n , i t i s not necessary. In these s t u d i e s , f o r ttV# and /hVd/ productions, with m u l t i p l e mixed speakers, the vowels were i d e n t i f i e d w e l l d e s p i t e the lac k of cues f o r n o r m a l i z a t i o n . It would seem that c a l i b r a t i o n of the vowel space may be accomp 1 i sh ed without t h e s e a l i n g i n format i on a v a i l a b l e i n m u l t i p l e vowel p r o d u c t i o n s by the same speaker- S c a l i n g may be accomplished with i n f o r m a t i o n from a s i n g l e vowel by employment of the v a l u e s o f the forrnants or t h e i r r e l a t i o n s between, as c i t e d e a r l i e r ( i . e . Neary, 1977; Skinner, 1977). Vowel forrnant n o r m a l i z a t i o n s t r a t e g i e s have had some success i n accounting f o r vowel i d e n t i f i c a t i o n over speakers with d i f f e r e n t s i z e v o c a l t r a c t s , but they s t i l l do not adequately account f o r l i s t e n e r success with vowels produced i n CVC context, where forrnants are s h i f t e d from the p o s i t i o n o c c u r r i n g when they are produced i n i s o l a t i o n ( c a n o n i c a l p o s i t i o n ) . ft p o s s i b i l i t y i s that the vowel or s y l l a b l e may be s u f f i c i e n t i n i t s e l f to provide vowel n o r m a l i z a t i o n i n f o r m a t i o n . T h i s i n f o r m a t i o n may be q u i t e d i f f e r e n t from the forrnant frequency v a l u e s u s u a l l y c o nsidered. P o s s i b l e cues i n c l u d e d u r a t i o n d i f f e r e n c e s , i n t e n s i t y d i f f e r e n c e s , degree of d i p h t h o n g i z a t i o n , and t r a n s i t i o n r a t e and d i r e c t ion. Vowel d u r a t i o n has been shown to separate vowels i n t o two groups; tense and lax, with tense vowels (/i/,/ae/,/a/,/3 / ,/u/) hciving longer s t e a d y - s t a t e segments ( L e h i s t e & Peterson, 1961) and lax vowels (/I/,/E/,/a/,/U/) having longer o f f g l i d e s . Lack of d u r a t i o n a l cues can have s i g n i f i c a n t l y d e t r i m e n t a l e f f e c t s on vowel i d e n t i f i c a t i o n as Fair b a n k s and Grubb (1961) found f o r s u s t a i n e d i s o l a t e d vowels. I n t e n s i t y a l s o v a r i e s s y s t e m a t i c a l l y among vowels. L e h i s t e and Peterson (1959) found a 5 dB range of i n t e n s i t y f o r vowels, with /a/ the most i n t e n s e and / i / the l e a s t . Vowels a l s o vary i n t h e i r degree of diphthongal movement, which can be used i n i d e n t i f i c a t i o n (flssrnan et. a l , 1982). The diphthongal o f f g l i d e s f o r the tense vowels tend toward the extremes of the vowel space ( / i / f o r f r o n t , /u/ f o r back). Conversely, f o r c e r t a i n North flmerican d i a l e c t s , lax vowels may show o f f g l i d e s i n the d i r e c t i o n of schwa, ac c o r d i n g t o Joos (1948). Although vowels produced i n i s o l a t i o n are c o n s i d e r e d the i d e a l or c a n o n i c a l forms, s e v e r a l s t u d i e s have found high e r r o r r a t e s f o r n a t u r a l l y spoken, i s o l a t e d vowels ( L e h i s t e & Meltzer, 1973; Strange, Verbrugge, Shankweiler & Edman, 1976) and f o r vowels excerpted from context (Bond, 1975; Fu.jimura &• O c h i a i , 1963), suggesting that more i n f o r m a t i o n than that contained i n the vowel s t e a d y - s t a t e may be u s e f u l i n vowel p e r c e p t i o n . In connected d i s c o u r s e , vowels occur j o i n e d t o consonants. The t r a n s i t i o n i n t e r v a l s o c c u r r i n g between the previous consonant t e r m i n a t i o n and the vowel s t e a d y — s t a t e , or between the vowel s t e a d y - s t a t e and the f o l l o w i n g consonant i n i t i a t i o n were i n i t i a l l y a s c r i b e d the r o l e of c h a r a c t e r i z i n g the consonants and were not c o n s i d e r e d u s e f u l f o r vowel i d e n t i f i c a t i o n s It i s assumed that such changes are cues f o r the p e r c e p t i o n of the consonants r a t h e r than l i n g u i s t i c a l l y s i g n i f i c a n t components of the v o c a l i c nucleus ( i n monophthongs). ( L e h i s t e & Peterson, 1961, p. £68) The c o n t r i b u t i o n of t r a n s i t i o n s t o vowel i dent i f i cat i on was f i r s t i n d i c a t e d by Lindblom and Studdert—Kennedy (1967). They s y n t h e s i z e d a vowel s t e a d y - s t a t e continuum v a r y i n g from L" 11 to CUIl,, These vowels were both preceded and followed by one of two consonantal frames, Ew--w3 or C j - j l l , and presented i n CVC form. For the same vowel midpoint frequency, presence of d i f f e r e n t consonant t r a n s i t i o n s s h i f t e d the category boundary between the vowels. T h i s i n d i c a t e d that the i d e n t i t y of a vowel i s determined not onl y by the forrnant— frequency p a t t e r n of the s t e a d y - s t a t e , but a l s o by the d i r e c t i o n and r a t e of adjacent "consonant" forrnant t r a n s i t i o n s . T r a n s i t i o n s i n c r e a s e d i n s i g n i f i c a n c e v i s - a - v i s vowel i d e n t i f i c a t i o n when b e t t e r vowel i d e n t i f i c a t i o n r a t e s were found to occur i n a consonantal context r a t h e r than i n i s o l a t i o n . M i l l a r and ft i nswort h (197£) r e p o r t ed a more r e l i a b l e and un i form i d e n t i f i c a t i o n of vowels s y n t h e s i z e d i n a /h-d/ context than when f o r the same vowels s y n t h e s i z e d as i s o l a t e d steady s t a t e s . T h i s f i n d i n g suggested, that, although /h-d/ i s sometimes co n s i d e r e d a n u l l context, because of i t s minimal e f f e c t on forrnant steady-s t a t e values, the t r a n s i t i o n s are somehow c o n t r i b u t i n g t o ident i f i c a t ion. More c o n v i n c i n g e;vidence cornes frorn cornparisons of natura 1 vowels spoken i n CVC co n t e x t s and i n i s o l a t i o n where both p r o d u c t i o n s have s p e a k e r - c o n t r o l l e d onsets and o f f s e t s . Strange et a l . (1976) compared p e r c e p t i o n of i s o l a t e d vowels t o vowels i n /pVp/ s y l l a b l e s spoken by men, women and c h i l d r e n . Speakers were given one p r a c t i c e t r i a l f o r each s t i m u l u s type. The recorded items were presented t o l i s t e n e r s i n s i n g l e and mixed speaker c o n d i t i o n s . S u b j e c t s heard the s t i m u l i over loudspeakers and responded on s c o r e s h e e t s coded with the o r t h o g r a p h i c e q u i v a l e n t of each vowel i n a "pVp" form. E r r o r s f o r the /pVp/ c o n d i t i o n s were 17. @°/. and 9.5"/. f o r mixed and blocked speakers r e s p e c t i v e l y . For the i s o l a t e d vowels, e r r o r s were 42.6% and 31.2/4. S i m i l a r r e s u l t s were found by G o t t f r i e d and Strange (1980) f o r /pVp/,/bVb/, and /kVk/, but not f o r /gVg/. These r e s u l t s show a very poor i d e n t i f i c a t i o n performance f o r i s o l a t e d vowel p r o d u c t i o n s and a s t r o n g advantage f o r vowels produced i n CVC c o n t e x t s as compared to vowels produced i n i s o l a t i o n . These s t u d i e s , showing b e t t e r i d e n t i f i c a t i o n performance f o r vowels produced i n CVC sequences as compared to that of vowels produced i n i s o l a t i o n , and high e r r o r r a t e s f o r i s o l a t e d vowel productions, were c r i t i c i z e d on procedural grounds. C r i t i c i s m s i n v o l v e d the dubious l i s t e n i n g q u a l i t y obtained i n f r e e f i e l d t e s t i n g , the lack of checks on vowel q u a l i t y , and lack of d i a l e c t matching. A major source of c r i t i c i s m was the use of CVC o r t h o g r a p h i e a l l y s p e l l e d response ch o i c e s , f o r both i s o l a t e d vowels and vowels i n CVC frames. It was f e l t t h i s i n troduced a response i n c o m p a t i b i l i t y f o r the i s o l a t e d vowels, thus d e p r e s s i n g performance i n that c o n d i t i o n . Assrnan et a l . (1982) performed two experiments to e v a l u a t e the e f f e c t o f response types on performance. They attempted to provide? high q u a l i t y l i s t e n i n g c o n d i t i o n s by use of headphones, monitoring of q u a l i t y , and d i a l e c t matching ( A l b e r t a , Canada). T h e i r f i r s t experiment examined whether a s i g n i f i c a n t p r o p o r t i o n of e r r o r s were due to m i s l a b e l l i n g r a t h e r than pe r c e p t u a l c o n f u s i o n s . S u b j e c t s were presented /pVp/ s t i m u l i . Responses were /hVd/ keywords and r e p e t i t i o n of the sound heard. E r r o r s i n the w r i t t e n c o n d i t i o n were 17"/. f o r #V# and 15% f o r /pVp/, thus showing no c o n t e x t u a l advantage. E r r o r s i n the w r i t t e n and spoken c o n d i t i o n were 5% f o r both #V# and /pVp/„ The marked decrease i n e r r o r s f o r the combined c o n d i t i o n as compared to the w r i t t e n c o n d i t i o n showed that a l a r g e p r o p o r t i o n of e r r o r s i n key word t a s k s can be the r e s u l t of m i s l a b e l l i n g r a t h e r than pe r c e p t u a l c o n f u s i o n s . In the second experiment, #V# and /pVp/ s t i m u l i were again presented and keyword (/hVd/) and s p e l l e d responses C/pVp/) were compared. The h y p o t h e s i s was that s p e l l e d responses are e a s i e r f o r l i s t e n e r s than keyword responses. E r r o r s were s i g n i f i c a n t l y d i f f e r e n t only f o r the /pVp/ s t i m u l i , with the keyword task producing more errors;;. Both these experiments showed response types can i n f l a t e e r r o r r a t e s and can do so se1ect i ve1y. A d d i t i o n a l s t u d i e s a l s o showed that when a t t e n t i o n i s paid t o obt a i n i n g q ua1i t y prod uct i ons and 1i st en i ng cond i t i ons, matching r e g i o n a l d i a l e c t s , and p r o v i d i n g compatible response a 11 e m a t i ves f o r bot h cond i t i ons, un i f or rn 1 y 1 ow e r r o r r a t es and l i t t l e c o n t e x t u a l advantage i s found (Kahn, 1978; Macchi, 1980? D i e h l , Buchwald, McCusker, & Chapman, 1981). Vowels produced i n C V C s are not c o n s i s t e n t l y e a s i e r t o i d e n t i f y than vowels produced i n i s o l a t i o n . However, performance on vowels produced i n C V C s i s no worse than performance on vowels produced i n i s o l a t i o n e i t h e r , which i s what might be expected i f the presence of t r a n s i t i o n s served only t o " d i s t o r t " vowels from t h e i r c a n o n i c a l s t e a d y - s t a t e f r e q u e n c i e s . Compelling evidence f o r the involvement of t r a n s i t i o n s i n the vowel i d e n t i f i c a t i o n process comes from two recent s t u d i e s , by Jenkins, Strange and Edman (1983) and by Strange, J e n k i n s and Johnson (1983). These s t u d i e s i n v e s t i g a t e d how str o n g a cue t r a n s i t i o n s are when no s t e a d y - s t a t e i n f o r m a t i o n i s provided. R e s u l t s were s i m i l a r f o r the two s t u d i e s . The l a t t e r study w i l l be d e s c r i b e d s i n e e i t c o n t a i n s a more systemat i c treatment of the var iab1es i n ques t i o n . Natural /bVb/ u t t e r a n c e s were modified by s u b s t i t u t i n g s i l e n t i n t e r v a l s f o r the t r a n s i t i o n s or the s t e a d y - s t a t e p o r t i o n s of the vowe 3.. Norma 1 d urat i on d i f f e r e n c e s were ma i nt a i ned t o v a r y i n g degrees: f o r the s t e a d y - s t a t e only s t i m u l i (SS), the remaining c e n t e r was e i t h e r kept at i t s n a t u r a l d u r a t i o n or was cut to a standard length; f o r the t r a n s i t i o n s only s t i m u l i (TR) , the s i l e n t c e n t e r was e i t h e r i t s n a t u r a l d u r a t i o n or was the standard length, while the t r a n s i t i o n s maintained t h e i r o r i g i n a l lengths. The r e s u l t s showed, f o r s t i m u l i with d u r a t i o n a l i n f o r m a t i o n (both the t r a n s i t i o n and s t e a d y — s t a t e d u r a t i o n s preserved), e r r o r r a t e s f o r both SS and TR s t i m u l i were very low; B% and &% r e s p e c t i v e l y . A d d i t i o n a l l y , performance was not s i g n i f i c a n t l y d i f f e r e n t between the CO and the TR s t i m u l i . Reduced d u r a t i o n i n f o r m a t i o n (standard length center, normal d u r a t i o n t r a n s i t i o n ) d i d not s i g n i f i c a n t l y a f f e c t the TR s t i m u l i . Absence of d u r a t i o n a l i n f o r m a t i o n (standard length center, no t r a n s i t i o n s ) d i d damage the SS sounds. There i s nob only an advantage i n the i d e n t i f i c a t i o n of vowels produced i n a CVC sequence, but t r a n s i t i o n s without the presence of s t e a d y - s t a t e or f u l l d u r a t i o n i n f o r m a t i o n a l s o p r o v i d e adequate i n f o r m a t i o n f o r vowel i d e n t i f i c a t i o n . Based on these f i n d i n g s , an approach i n c o r p o r a t i n g dynamic p r o c e s s i n g i n t o vowel p e r c e p t i o n was put forward (Strange et a l . , 1976; Jenkins et a l . , 1983; Strange et a l . , 1983) as an a l t e r n a t i v e to s t a t i c vowel space c h a r a c t e r i z a t i o n s . The dynamic process d e s c r i b e d i d e n t i f i c a t i o n of a vowel as i n v o l v i n g both temporal and s p e c t r a l i n f o r m a t i o n over the course of a complete CVC s y l l a b l e . I d e n t i f i c a t i o n c o u l d be accomplished through the p a r t i a l i n f o r m a t i o n presented i n i s o l a t e d vowel productions, such as the s t e a d y - s t a t e forrnant frequency values, but normal i d e n t i f i c a t i o n i n v o l v e d i n f o r m a t i o n i n t e g r a t e d over an e n t i r e CVC time period. T h i s dynamic process approach i s r e m i n i s c e n t of an a c o u s t i c model developed by Stevens, House and Paul (1966) and based on a dynamic a r t i c u l a t o r y d e s c r i p t ion. These authors d e s c r i b e d a r t i c u l a t i on o f a CVC s y l i a b l e as cons i s t i n g o f a s uper pos i t i on of a r t i c u l a t o r y events s e q u e n t i a l l y ordered but o v e r l a p p i n g . Pis a r e s u l t , "the a r t i c u l a t o r y a c t i v i t y - and the r e s u l t i n g a c o u s t i c output - that c h a r a c t e r i z e s a s y l l a b i c nucleus depends on the i n i t i a l and f i n a l consonants as well as on the vowel i t s e l f " ( i b i d , p. 1S3). T h e i r dynamic model c a l c u l a t e d v o c a l i c c ontours r e p r e s e n t i n g the f i r s t and second forrnants as a f u n c t i o n of t i me. The pararnet e r s used i nc 1 uded i n i t :i. a 1 and f i na 1 f r e q u e n c i e s , midpoint f r e q u e n c i e s , d u r a t i o n s , and measures of t rans i t ion curvat ure. Strange and her c o l l e a g u e s (Strange et a l . , 1976; Strange et al„, 1983; J e n k i n s et a l . , 1983) d i d not s p e c i f y the d e t a i l s of t h e i r model i n p e r c e p t u a l terms, taut a f a i r l y s u c c e s s f u l attempt to i n v o l v e dynamic v a r i a b l e s was made by flssman et a l . (1982), who developed t h i s dynamic process approach f u r t h e r by combining measures of " s t e a d y - s t a t e i n f o r m a t i o n " (forrnant f r e q u e n c i e s and fundamental frequency), "dynamic i n f o r m a t i o n " (forrnant t r a n s i t i o n s l o p e s and duration) and "speaker i n f o r m a t i o n " (VT length) to achieve a c h a r a c t e r i z a t i o n that r e s u l t e d i n r e l a t i v e l y l i t t l e o v e r l a p between vowel c a t e g o r i e s . 2. 3 Coart i cu1at ion Evidence suggests that a s t a t i c c o n c e p t u a l i z a t i o n of vowel i dent i f i c a t i on, i nvo 1 v i n g d e r i v a t i on of forrnant frequency va 1 ues, i s i n s u f f i c i e n t t o adequately d e s c r i b e vowel p e r c e p t i o n . A dynamic approach, i n v o l v i n g t i m e — v a r y i n g as well as s t a t i o n a r y cues, seems t o account f o r the process of i d e n t i f i c a t i o n b e t t e r . I f a dynamic model i s to be p o s t u l a t e d , then the presence and u t i l i t y of phonetic o v e r l a p extending f u r t h e r than that o c c u r r i n g f o r vowels produced i n CVC sequences should be s p e c i f i e d . How d i s t a n t can the cues to phoneme i d e n t i t y be? Is the s y l l a b l e an adequate u n i t of a n a l y s i s ? Both segmentation s t u d i e s and c o a r t i cu1at i on st ud i es suggest ot herw i se. Normal d i s c o u r s e i n v o l v e s the blending of vowels and consonants to such a degree that i t i s d i f f i c u l t i f not impossible t o segment the a c o u s t i c s i g n a l i n t o d i s c r e t e segments, as Liberman, Cooper, Shankweiler and Studdert-Kennedy (1967) point out i n t h e i r review of work i n speech p e r c e p t i o n . The s e n t e n t i a l context of connected d i s c o u r s e p r o v i d e s cues not only at h i g h e r l e v e l s of a n a l y s i s such as s y n t a c t i c and semantic, but at the r e l a t i v e l y p r i m i t i v e phonetic l e v e l as w e l l . Ladefoged and Broadbent (1957), as d e s c r i b e d e a r l i e r , showed that i d e n t i f i c a t i o n o f a word i s g r e a t l y i n f l u e n c e d by the range of v a r i a t i o n of each forrnant frequency i n the preceding c a r r i e r phrase. Verbrugge, Strange, Shankweiler, & Edman (1976) compared the i d e n t i f i c a t i o n o f three types of s t i m u l i ; (1) s y l l a b l e s i n t he i r c a r r i er sent ences; (£) s y l l a b l e s exc i sed from the same ut t e r a n c e ; (3) e x c i s e d s y l l a b l e s preceded by p r e c u r s o r s t r i n g s with c o n f l i c t i n g tempo i n f o r m a t i o n . E r r o r s i n c r e a s e d from (1) to (3) and i t was concluded that sentence context a l l o w s adjustment t o speaker tempo, thus a i d i n g i n vowel i d e n t i f i c a t i o n . Connected d i s c o u r s e not only p r o v i d e s a d d i t i o n a l i n f o r m a t i o n , but, by the very nature of i t s o v e r l a p p i n g connect i ons and para11e1 t ransmi ss i on, d i s t r i b u t e s ph one i d e n t i f i c a t i o n i n f o r m a t i o n over a time l a r g e r than the d u r a t i o n of the heard phone, thus a f f e c t i n g the r e a l i z a t i o n of the u n d e r l y i n g phonemes and changing thern from what they are or would appear as i n i s o l a t i o n . The process i n v o l v i n g t h i s spread of i n f o r m a t i o n i s c a l l e d c o a r t i c u l a t i o n and occurs because of the programming and e x e c u t i o n p r o p e r t i e s of the a r t i c u l a t o r y mechanism. C o a r t i c u l a t o r y e f f e c t s are of two types, r e f e r r e d to as ant i c i pat ory and c a r r y o v e r . Le f t - 1 o - r i gh t or p o s t - a r t i c u1at ory e f f e c t s , termed c a r r y o v e r c o a r t i c u 1 a t i o n , e f f e c t s occur when the a r t i c u l a t i o n of a phone i s i n f l u e n c e d by the a r t i c u l a t i o n of a prev i o u s phone. Carryover c o a r t i c u l a t i o n i s the most p r e d i c t a b l e of the two types, being p r i m a r i l y caused at the exe c u t i o n l e v e l by rnechano i n e r t i a l f a c t o r s s p e c i f i c t o the a r t i c u l a t o r s . A n t i c i p a t o r y e f f e c t s are r i g h t - t o - l e f t or p r e - a r t i c u l a t o r y 5 the a r t i c u l a t i o n of a phone i s i n f l u e n c e d by that of a subsequent phone. A n t i c i p a t o r y c o a r t i c u l a t i o n o r i g i n a t e s at the planning l e v e l , because the speech gestures are g e n e r a l l y not independent, nor s t r i c t l y l i n e a r l y sequenced. A n t i c i p a t o r y e f f e c t s are of gr e a t e r i n t e r e s t than c a r r y o v e r e f f e c t s i n models of speech production, although both are of concern i n p e r c e p t i o n . Kozhevnikov and C h i s t o v i c h (1965), i n t h e i r model of speech prod uct i on, post u1at ed an "a r t i c u l a t ory s y l l a b l e " i n wh i ch commands f o r the? e n t i r e a r t i c u l a t o r y s y l l a b l e are i n i t i a t e d s i m u l t a n e o u s l y . The commands are a l s o executed s i m u l t a n e o u s l y i f the requirements are non-competing, but s e q u e n t i a l l y i f they are competing. For Russian, c o a r t i c u l a t i o n i s con s i d e r e d t o be maximum over the s t r o n g l y c o h e s i v e CV (or CCV,...) s y l l a b l e u nit and minimum a c r o s s s y l l a b l e boundaries. The model was based on t h e i r (1965) study of electromyographic a c t i v i t y i n the o r b i c u l a r i s o r i s muscle, f o r l i p r o u n d i n g , during p r o d u c t i o n of v a r i o u s vowel and consonant combinations. Fromkin (1966) a l s o d i d an EMG study o f the muscle used i n 1iprounding during rounded and unrounded vowel and /b, p/ pr o d u c t i o n s i n CVC sequences. She found that d i f f e r e n t motor commands produce d i f f e r e n t muscular gestures f o r i n i t i a l /b/ or /p/ as compared t o f i n a l , but vowel context d i d not s i g n i f i c a n t l y a f f e c t EMC3 a c t i v i t y , nor d i d consonant context seem to a f f e c t EMG a c t i v i t y i n vowels. Fromkin suggested that some asp e c t s of context somehow r e s t r i c t or r e o r g a n i z e the neuromuscular commands and gestures f o r some phonemes, but t h i s e f f e c t i s not common to a l l c h a r a c t e r i s t i c s of context. One e x p l a n a t i o n given was the p o s s i b l e e x i s t e n c e of a feedback system concerning the s t a t e of muscle a c t i v i t y , wherein the b a s i c unit of speech p r o d u c t i o n i s s t i l l a phoneme-size u n i t . Another e x p l a n a t i o n was that the minimal l i n g u i s t i c u n it at the motor command l e v e l may be l a r g e r than the phoneme, p o s s i b l y on the order of a s y l l a b l e . Henke's (1966) model gave support f o r a phoneme-sized e x e c u t i o n u n i t , but added a "look ahead" system. In h i s model, pr o d u c t i o n i s organized phoneme by phoneme, but i n c l u d e s scanning for| upcoming f e a t u r e s . I f a phoneme c u r r e n t l y being r e a l i z e d i s unmarked f o r a p a r t i c u l a r f e a t u r e , the f e a t u r e w i l l be set to the val u e of the next occurrence where i t i s marked. T h i s value w i l l be determined by means of the look-ahead mechanism, which scans upcoming phonemes to examine the value of each f e a t u r e . Ohrnan <1966) a l s o espoused a scanning approach. He conceived of phoneme pr o d u c t i o n i n connected d i s c o u r s e as i n v o l v i n g i n v a r i a n t i n t e n t i o n s , with v a r i a b l e motoric r e a l i z a t i o n s . These r e s u l t i n v a r i a b l e VT shapes, which are the r e s u l t of l a t e t e r m i n a t i o n of the previous phoneme (due to i n e r t i a l f a c t o r s ) and e a r l y i n i t i a t i o n of the upcoming phoneme, both o c c u r r i n g during the e x e c u t i o n of the c u r r e n t phoneme. Planning over at l e a s t the i n t e r v a l from one vowel t o the next over i n t e r v e n i n g consonants i s thought to occur. tinman d e s c r i b e s the a r t i c u l a t i o n o f each phoneme of a VCV sequence as i n v o l v i n g separate but o v e r l a p p i n g s e t s of muscles, each set having a sepa r a t e n e u r a l r e p r e s e n t a t i o n i n the c o n t r o l networks of the b r a i n . The vo c a l t r a c t shape at any point i n time i s a f u n c t i o n of messages from m u l t i p l e channels. Although the phoneme c u r r e n t l y i n pr o d u c t i o n i s the? main i n f o r m a t i o n c o n t r i b u t o r t o these channels, ... a VCV ut t e r a n c e of the kind s t u d i e d here can, a c c o r d i n g l y , not be regarded as a l i n e a r sequence of th r e e s u c c e s s i v e gestures. We have c l e a r evidence that the stop-consonant gestures are a c t u a l l y superimposed on a context-dependent vowel s u b s t r a t e that i s present during a l l of the consonantal gesture. (Ohrnan, 1966, p. 165) Ohrnan (1967) matched l a t e r a l x-ray data of VCV pr o d u c t i o n s t o h i s numerical model of c o a r t i c u l a t i o n . He found that a c t u a l v o c a l t r a c t shapes f o r Swedish VCV u t t e r a n c e s compared well with shapes generated by h i s c o a r t i c u l a t i o n formula. His model was based on spect rograph i c dat a c o n c e r n i ng t ransconsonant a1 c o a r t i cu1at ion r e p o r t e d i n Ohrnan (1966). T h i s data were obtained from American, Swedish and Russian speakers producing V tCV a u t t e r a n c e s . T r a n s i t i o n onset frequency and s t e a d y - s t a t e frequency were measured f o r each vowel. A measurable c a r r y o v e r c o a r t i c u l a t o r y e f f e c t from Vt on the t r a n s i t i o n p o r t i o n of the CV a and a s i m i l a r a n t i c i p a t o r y c o a r t i c u l a t o r y e f f e c t from V e on the t r a n s i t i o n p o r t i o n of Vt C were found. The e f f e c t on the vowel s t e a d y - s t a t e p o r t i o n was r e p o r t e d to be s m a l l . Major evidence f o r a n t i c i p a t o r y c o a r t i c u l a t i o n i n the speech p r o d u c t i o n process comes from a r t i c u l a t i o n s t u d i e s , e.g. fihman (1966; 1967). Carney and Moll (1971) extended these f i n d i n g s with a c i n e f 1 u o r o g r a p h i c i n v e s t i g a t i o n of f r i c a t i v e consonants i n VCV context. They found c o a r t i c u l a t i o n e f f e c t s s i m i l a r to those found by Ohman (1967). C o a r t i c u l a t o r y e f f e c t s have a l s o been i n v e s t i g a t e d f o r v e l a r movements by Moll and D a n i l o f f (1971). Four s u b j e c t s produced E n g l i s h sentences c o n t a i n i n g v a r i o u s combinations of nasal consonants (N), consonants (C), and vowels (V). These were examined c i n e f 1 u o r o g r a p h i c a l l y . The r e s u l t s i n d i c a t e d e x t e n s i v e a n t i c i p a t o r y c o a r t i c u l a t ion of v e l a r movement i n CVN and CWN sequences, beginning dur i n g the approach to the i n i t i a l vowel. C o a r t i c u l a t i o n e f f e c t s a c r o s s longer segments have a l s o been shown to occur (Benguerel and Cowan, 1974). Benguerel and Cowan examined the t i m i n g of upper l i p p r o t r u s i o n i n French, p a r t i c u l a r i l y the onset of p r o t r u s i o n i n a consonant c l u s t e r followed by a rounded vowel (e.g. / k s t r y / i n " l a dextre u n i v e r s e ! l e " ) . R e s u l t s showed that rounding movement cou l d s t a r t as e a r l y as f o u r t o s i x consonants before the vowel. The s t u d i e s c i t e d show that c o a r t i c u l a t i o n can be observed at bob h t h e a r t i c u1at ory and the a c o u s t i c 1eve1 t o vary i n g degrees depending on the f e a t u r e i n v o l v e d . The models of speech p r o d u c t i o n de?scri bed d i f f e r i n the degree? to which they admit c o a r t i c u l a t i o n , but a l l those c i t e d o f f e r p l a u s i b l e e x p l a n a t i o n s f o r c o a r t i c u1at i on i n prod uct i on o f s peech. £.4 P e r c e p t u a l E l f f e c t s of Coart i c u l a t ion C o a r t i c u l a t i o n i s a c o u s t i c a l l y present, but whether i t i s p e r c e i v a b l e , and i f and how i t i s i n v o l v e d i n the normal process of phoneme i d e n t i f i c a t i o n i s s t i l l to be determined. P e r c e p t i o n of c o a r t i c u l a t i o n c o u l d occur i n a number of ways. A s t r o n g c o a r t i c u l a t o r y presence would allow p r e d i c t i o n of the i d e n t i t y of an upcoming phone. Pi weaker m a n i f e s t a t i o n would be s u f f i c i e n t t o allow p r e d i c t i o n of f e a t u r e s (e.g. 1iprounding). Another p o s s i b i l i t y i s that c o a r t i c u l a t i o n may a f f e c t the nature of the phoneme i t o v e r l a y s , r e s u l t i n g i n a change i n i d e n t i t y or at l e a s t a change i n q u a l i t y . A l i , G a l l a g h e r , G o l d s t e i n and D a n i l o f f (1971) t e s t e d the perce i vabi .1 i t y of the nasa 1 coart i c u l a t i on descr i bed by Mo 11 and D a n i l o f f (1971). They s p l i c e d out the f i n a l consonant ( i n c l u d i n g i t s VC t r a n s i t i o n ) from E n g l i s h CVC and CVVC utterances, where the f i n a l C was sometimes a nasal consonant, and sometimes a non-nasal consonant. ££ s u b j e c t s were asked to i d e n t i f y whether the missing consonant was nasal or non—nasal. Nasal consonants were p r e d i c t e d c o r r e c t l y at b e t t e r than chance l e v e l s (p<.001). Stops were pe r c e i v e d as n a s a l s more f r e q u e n t l y than f r i c a t i v e s . Consonants f o l l o w i n g low vowels were perceived as n a s a l s more o f t e n than those f o l l o w i n g high vowels. A l i et a l . r e p o r t e d a l l of these f i n d i n g s to be c o n s i s t e n t with p r e v i o u s s t u d i e s on nasal i d e n t i f i c a t i o n such as L i n t z & Sherman (1961). Benguerel and Ode1man (1977) examined the pe r c e p t u a l s i g n i f i c a n c e of l i p rounding c o a r t i c u l a t i o n i n French vowels d e s c r i b e d by Benguerel and Cowan (197*4). They tru n c a t e d -CCCCV ut t e r a n c e s (or p a r t s t h e r e o f ) at fo u r d i f f e r e n t p o i n t s before the vowel and had French and E n g l i s h s u b j e c t s p r e d i c t the i d e n t i t y of the missing vowel. French and E n g l i s h s u b j e c t s were used to examine the p o s s i b i l i t y of d i f f e r e n t i a l s e n s i t i v i t y r e l a t e d t o l i n g u i s t i c e x p e r i e n c e s i n c e i n French, l i p r o u n d i n g on vowels i s c o n t r a s t i v e , while i n E n g l i s h i t i s not. The r e s u l t s showed that when segments up to and i n c l u d i n g at l e a s t h a l f the f i n a l consonant of the c l u s t e r are present ( i . e . t r u n c a t i o n occurs before the CV t r a n s i t i o n ) , s u b j e c t s c o r r e c t l y i d e n t i f i e d the missing vowel well above chance l e v e l . There were no s i g n i f i c a n t d i f f e r e n c e s between the French and E n g l i s h s u b j e c t s , suggesting that t h i s c o a r t i c u l a t o r y f e a t u r e can be perce i v e d s u b phonern i ca 11 y. Another study l o o k i n g at the perc e p t u a l e f f e c t s of l i p r o u n d i n g and n a s a l i t y was that of Sharf and O s t r e i c h e r (1973). They c o n s t r u c t e d u t t e r a n c e s of the form t3NC a-sV where M i s a nasal consonant (/n,m,Q/), C a. £' i s a consonant c l u s t e r c o n s i s t i n g of zero, one or two non-nasal consonants and V i s an unrounded or a rounded vowel (/i,u/)(e.g. /t^mi/, / t ^ r n t i / , / t a r n s t i / ) . The pos t n a s a l segment was then s p l i c e d out of the utterance. 37 s u b j e c t s were asked t o i d e n t i f y the nasal consonant, i n n o i s e and i n s i l e n c e , and the f i n a l (deleted) vowel i n s i l e n c e . I d e n t i f i c a t i o n of the nasal consonant was b e t t e r when a l l the postnas a l sounds were r e t a i n e d than when they were d e l e t e d , s u g g e s t i n g that the c a r r y o v e r c o a r t i c u 1 a t i o n normally present i n the second s y l l a b l e a i d s i d e n t i f i c a t i o n . When no consonant intervened betwee?n the nasal and the? vowel, i d e n t i f i c a t i o n of the missing vowel was s i g n i f i c a n t l y b e t t e r than chance?, demonstrating the presence and u t i l i t y o f a n t i c i pat ory c o a r t i cu1at i on. The-? consequences, at the pe r c e p t u a l l e v e l , of Ohrnan's (1366) f i n d i n g s on the a c o u s t i c presence of t r a n s c o n s o n a n t a l c o a r t i c u l a t i o n were examined by L e h i s t e and Shockey (1972). VCV utte r a n c e s , composed of the consonants /p/,/t/,/k/ and the vowels / i / , / a e / , / a / , and /u/, were cut i n two part s , at the v o i c e l e s s p l o s i v e gap, l e a v i n g e i t h e r a VC- or a -CV s y l l a b l e . S u b j e c t s were asked t o i d e n t i f y , from a c h o i c e of four vowels (/i/,/ae/,/a/,/u/) , the missing vowel. The r e s u l t s showed that the remaining i n f o r m a t i o n was i n s u f f i c i e n t t o cue the i d e n t i t y o f the missing vowel and responses d i d not f a l l i n t o c l a s s e s s h a r i n g some? f e a t u r e with the c o r r e c t response, such as high/low or front/back. C l a r k and Sharf (1973) examined the c o a r t i c u l a t i o n e f f e c t s on short-term memory r e c a l l . V 4CV a, V 4C- (V a d e l e t e d ) , and V tC ut t e r a n c e s were composed, where Vt was one of s i x lax vowels, C was a / t / , and V a was an /a/ or an / I / . These u t t e r a n c e s were presented t o s u b j e c t s who were asked t o r e c a l l the f i r s t vowel. The V i C - s t i m u l i r e s u l t e d i n b e t t e r r e c a l l s c o r e s than the other two c o n d i t i o n s only when the missing vowel (V a ) was the same as the vowel present ( V t ) . The authors concluded that the presence of c o a r t i c u l a t i o n i n f l u e n c e d r e c a l l under memory-taxing cond i t i ons: coart i cu1at ion of the d e l e t e d vowe1 appeared t o f a c i l i t a t e r e c a l l of V i i f V B was the same vowel, and to h i n d e r r e c a l l i f i t were d i f f e r e n t . The lack of f a c i l i t a t i o n i n the Vi CV a s t i m u l i was a t t r i b u t e d to p e r c e p t u a l o v e r l o a d from hearing two vowels. C o a r t i c u l a t i o n e f f e c t s are not s u f f i c i e n t t o p r e d i c t phoneme i d e n t i t y . They are s u f f i c i e n t t o p r e d i c t c e r t a i n phonetic f e a t u r e s , both c o n t r a s t i v e and n o n - c o n t r a s t i v e , such as rounding and n a s a l i t y , but not others, such as vowel height. The presence of c o a r t i c u l a t i o n can a i d i d e n t i f i c a t i o n and r e c a l l , although the presence of the u n d e r l y i n g elements can be d i s t r a c t i n g . Pm aspect of c o a r t i c u l a t i o n that has not been examined i s i t s e f f e c t on the u n d e r l y i n g phoneme. Does the presence of c o a r t i c u 1 a t i o n change the way i n which an a f f e c t e d phoneme i s perceived? PI p o s s i b l e candidate f o r t h i s i s the i d e n t i f i c a t i o n of vowels i n the presence of the t r a n s c o n s o n a n t a l c o a r t i c u l a t i o n d e s c r i b e d by Ohrnan (1966). •hman r e p o r t e d that, i n VCVs, t r a n s i t i o n s were a f f e c t e d to a much g r e a t e r degree than s t e a d y - s t a t e s by c o a r t i c u 1 a t i o n from the t r a n s c o n s o n a n t a l vowel. The recent work by J e n k i n s et a l . (1983) and Strange et a l . (1983) suggest that, i n a CVC utterance, t r a n s i t i o n s c a r r y s u f f i c i e n t i n f o r m a t i o n to allow vowe1 id ent i f i cat i on s c o r e s com parable t o t hose f o r the ent i r e CVC. N e i t h e r Strange et a l . (1983), nor Jenkins et a l . (1983) e x c i s e d the experimental s t i m u l i out of a l a r g e r utterance, consequently, the e f f e c t of non-adjacent elements could not be examined. It i s a well supported f a c t that items removed from context are l e s s i n t e l l i g i b l e than i s o l a t e d productions, due to l o s t i n f o r m a t i o n about context and r a t e , but the e f f e c t of the c o a r t i c u l a t i o n r e s u l t i n g from producing them i n context i s not u s u a l l y c o n sidered. The r e l a t i v e importance o f the v a r i o u s cues used i n vowel i d e n t i f i c a t i o n , as determined when produced i n i s o l a t i o n , may change when the same vowels are taken out of a l a r g e r context, where the f o r c e s of c o a r t i c u l a t i o n s e l e c t i v e l y a f f e c t a s p e c t s of these cues. PI I MS OF THE EXPERIMENT Cues t o vowel p e r c e p t i o n i n c l u d e forrnant f r e q u e n c i e s , d urat i on, d i phthonga 1 movement and consonant --vowe 1 t r a n s i t i ons. Recent work (Strange et a l . , 1983; Jen k i n s et a l . , 1983) has suggested that consonant-vowel t r a n s i t i o n s alone c o n t a i n s u f f i c i e n t i n f o r m a t i o n t o permit vowel i d e n t i f i c a t i o n . The speech segments used by these authors have been produced as CVC s y l l a b l e s and have not been s p l i c e d out from longer u t t e r a n c e s . S t u d i e s have shown that the surrounding phonetic context i n f .1 uences the nat ur e o f the consonant - vowe 1 un i t. T h i s i n f l u e n c e i s str o n g enough t o permit i d e n t i f i c a t i o n of c e r t a i n f e a t u r e s of the source o f i n f l u e n c e . A c c o r d i n g l y , i t s e f f e c t on the i n t e l l i g i b i l i t y of the u n d e r l y i n g segment could be expected t o be reasonably strong, at l e a s t i n some s i t u a t i o n s . S i n c e t h i s e f f e c t has been shown to d i f f e r e n t i a l l y a l t e r p a r t s of the vowel, the q u e s t i o n asked w i l l be whether i t a l t e r s the r e l a t i v e u t i l i t y o f the v a r i o u s cues t o vowel i d e n t i f i c a t i o n . T h i s study examines the e f f e c t of context on the cues t o vowel p e r c e p t i o n . The c o n t e x t u a l e f f e c t t o be examined i s tr a n s c o n s o n a n t a l c o a r t i c u l a t i o n , which i s known to a f f e c t the t r a n s i t i o n part of the vowel more than the s t e a d y - s t a t e part (Ohrnan, 1966) . It i s expected that the changes t o the t r a n s i t i o n s caused by tr a n s c o n s o n a n t a l c o a r t i c u l a t i o n w i l l reduce the i n t e l l i g i b i l i t y of u t t e r a n c e s which have t r a n s i t ions as the primary cues ( i . e . the s t e a d y - s t a t e p o r t i o n of the vowel removed). The degree of i n t e l l i g i b i l i t y should be l e s s than that of vowels c o n t a i n i n g s t e a d y - s t a t e and t r a n s i t i o n i n f ormation, or st eady-st at e i nformat i on a1one. The degree of i nt e11i g i b i 1 i t y w i l l be measured i n a f o r c e d c h o i c e i d e n t i f i c a t i o n task. It i s expected that c o r r e c t i d e n t i f i c a t i o n r a t e s w i l l be lower f o r the t r a n s i t i o n o n l y s t i m u l i s p l i c e d out of longer u t t e r a n c e s than f o r unmodified s t i m u l i , or s t e a d y - s t a t e only s t i m u l i , both s p l i c e d out of longer u t t e r a n c e s . S t a r t i n g from recorded u t t e r a n c e s of the form V tCV f iCV 4, the outer vowels are removed and the sounds a l t e r e d f o l l o w i n g the Strange et a l . (1383) procedure f o r c r e a t i n g s t e a d y - s t a t e only and t rans i t i on—on .1 y st i rnu 1 i . The ex per i merit t o be descr i bed examines (1) whether consonant-vowel t r a n s i t i o n s are s u f f i c i e n t t o cue vowel i d e n t i f i c a t i o n i n the presence of c o a r t i c u l a t o r y e f f e c t s . (£') whether d i f f e r e n t vowel c o n t e x t s d i f f e r e n t i a l l y a f f e c t vowel cues ( s t e a d y — s t a t e and t r a n s i t i o n s ) . £8 Chapter 4 METHODOLOGY 4.1 P r e p a r a t i o n o f Test Tapes Speech M a t e r i a l s The speech m a t e r i a l s from which the s t i m u l i were e x t r a c t e d were n a t u r a l u t t e r a n c e s of the form V | C V 8 C V i . The consonant (C) was always the voi c e d p l o s i v e /b/. V i was one of the three vowels, /a/, / i / , /u/, a l l found by Ohrnan (1966) t o have a tra n s c o n s o n a n t a l c o a r t i c u l a t i o n e f f e c t . V a was one of ten v o c a l i c n u c l e i ; / i / 5 / I / , / e l / , /E/, /ae/, /a/, /'"'/, /oU/, /U/, /u/. fi male adult speaker from the same r e g i o n as the s u b j e c t s (Western Canada), with some phonetics experience, produced the utt e r a n c e s . He read them from a l i s t w r i t t e n i n phonetic symbols, with c a r e f u l monitoring t o ensure c o r r e c t p r o n u n c i a t i o n . I n s t r u c t i o n s t o the speaker were that he say each u t t e r a n c e with even s t r e s s on each s y l l a b l e or, f a i l i n g t h a t , with s l i g h t l y i n c r e a s e d c e n t e r s y l l a b l e s t r e s s , at a comfortable r a t e . He could p r a c t i c e beforehand, repeat the utterance, and pause as needed. Each token was recorded a number of times. The tokens s e l e c t e d f o r f u r t h e r e v a l u a t i o n were chosen f o r the accuracy of production, r e l a t i v e evenness of p i t c h and s t r e s s , and lack of extraneous n o i s e such as paper r a t t l i n g . The tokens were then t r a n s c r i b e d by thr e e l i s t e n e r s t r a i n e d i n t r a n s c r i p t i o n but uninformed as to the nature of the experiment. They were t o l d that the tokens were of the form V^CVBCVI (see above). They were encouraged t o t r a n s c r i b e using whatever l a b e l s they f e l t necessary. They were a l s o asked to make q u a l i t a t i v e comments on p i t c h , s t r e s s and noise, whenever p o s s i b l e . Ed i t ing E d i t i n g was done on a PDP-12 computer with WAVES, a set of programs w r i t t e n by L l o y d Rice at UCLA. The audiotaped speech was played back at haIf-speed, low pass f i l t e r e d at 2. 5 kHz and t r a n s f e r r e d t o the PDP-12 computer and d i g i t i z e d with a 10-bit a n a l o g - t o - d i g i t a l c o n v e r t e r at 6 kHz sampling frequency. The V »C V B C V i s were reduced t o - C V 8 C - ' s and th r e e s e t s of s t i m u l i were generated from these with WPVES. The f i r s t set of s t i m u l i c o n s i s t e d of the u n a l t e r e d -CV a C-'s ( c o n t r o l or CO c o n d i t i o n ) , the second set had the t r a n s i t i o n s r e p l a c e d with s i l e n c e (Steady-State Only or SS c o n d i t i o n ) and t h i r d set had the s t e a d y — s t a t e p o r t i o n r e p l a c e d with s i l e n c e ( T r a n s i t i o n s On1y or TR cond i t i on). The o u t p o i n t s f o r e l i m i n a t i o n of the outer vowels were determined on spectrograms and c o r r o b o r a t e d with v o i c i n g i n f o r m a t i o n obtained from the s p e e c h - s i g n a l d i s p l a y used f o r e d i t i n g . The boundary of the segment r e t a i n e d (-CVeC-) was cons i d e r e d t o be from the i n i t i a l consonant r e l e a s e t o the f i n a l consonant c l o s u r e . The c r i t e r i a used i n s e l e c t i n g cut p o i n t s were based on the procedure used by Strange et a l . (1983). They r e p o r t e d t h r e e s e t s of cut p o i n t s , a r r i v e d at by examination of L e h i s t e and Peterson's (1961) data on vowel t r a n s i t i o n and c e n t e r durat ions. In L e h i s t e and Peterson (1961), t h e r e were two groups of simple s y l l a b i c n u c l e i ; short and long. The two types were d i f f e r e n t i a b l e , not by a b s o l u t e d u r a t i o n , but by the r e l a t i v e d u r a t i o n of t h e i r c e n t e r s and o f f g l i d e s . Long vowels had longer c e n t e r s and s h o r t e r o f f g l i d e s , and short vowels had s h o r t e r c e n t e r s and longer o f f g l i d e s . t h i r d group of s i n g l e v o c a l i c n u c l e i , composed of / e i / and /ou/, were d e s c r i b e d as long, complex n u c l e i , with an extremely long o n g l i d e / e i / or o f f g l i d e /o u/. Strange et a l . (1983) used a number of d i f f e r e n t e u t p o i n t s f o r the o n g l i d e and o f f g l i d e s . They set the i n i t i a l p o r t i o n , or o n g l i d e , at a constant 15°/. of the t o t a l d u r a t i o n f o r a l l vowel types. For the o f f g l i d e p o r t i o n , they used t h r e e e u t p o i n t s . Long vowels were cut at the l a s t £IZI°/. and short vowels at the l a s t 35°/.. The t h i r d c u t p o i n t , f o r a group termed the i n t e r m e d i a t e vowels, / i / and /u/, was at B5% of the t o t a l d u r a t i o n . There i s no d i s c e m a b l e reason why t h i s group was c r e a t e d ; / i / and /u/ are not i n t e r m e d i a t e i n o f f g l i d e length r e l a t i v e to t o t a l d u r a t i o n ( L e h i s t e & Peterson, 1961), nor i n a b s o l u t e d u r a t i o n (Peterson & L e h i s t e , I960). It i s p o s s i b l e that the goal of the m a n i p u l a t i o n was to prevent s t e a d y — s t a t e i n f o r m a t i o n from o c c u r r i n g i n the t r a n s i t i o n s , due to the i n d i v i d u a l v a r i a t i o n o c c u r r i n g f o r the p a r t i c u l a r vowels used by Strange et a l . (1983), but t h i s i s not a c t u a l l y s p e c i f i e d i n the paper. The cut p o i n t s f o r the s t i m u l i of the present experiment were a l s o based on L e h i s t e and Peterson (1981). The steady-s t a t e p o r t i o n was d e f i n e d as that time p e r i o d d u r i n g which the forrnants (I-1, FS, F"3) are steady ( i . e . p a r a l l e l t o the time a x i s on a spectrogram). T r a n s i t i o n s were measured from the onset of v o i c i n g t o the beginning of s t e a d y - s t a t e and from the end of s t e a d y - s t a t e to the c o n c l u s i o n of v o i c i n g at the point of consonant c l o s u r e . When the cut p o i n t s d e s c r i b e d above were a p p l i e d to the vowels f o r the present experiment, they d i d not match the a c t u a l s p e c t r a l boundary between the t r a n s i t i o n and s t e a d y - s t a t e . The g r e a t e s t concern was f o r those vowels where the r e l a t i v e length of t r a n s i t i o n s was s h o r t e r than the length s p e c i f i e d by Strange et a l . (1983). In those vowels, steady-s t a t e i n format i on i s present i n the t rans i t i on sect i ons, t h us p r o v i d i n g i n f o r m a t i o n known t o be u s e f u l i n i d e n t i f i c a t i o n ( i . e . t a r g e t i n f o r m a t i o n ) . The? vowels causing the major concern were /u/, /U/, /oU/, and /"'/. These vowels had near-h o r i z o n t a l t r a n s i t i o n s f o r e i t h e r or both the o n g l i d e and the o f f g l i d e . fls w e l l , the vowels / e l / and /oU/ presented the added problem of being placed i n the simple n u c l e i , long vowel category, although t h e i r g l i d i n g nature d i d not conform to the c u t p o i n t s f o r t h i s category. F i n a l l y , v a r i a t i o n i n t o t a l vowel, s t e a d y - s t a t e and t r a n s i t i o n d u r a t i o n s was much g r e a t e r i n t h i s experiment than i n Strange et a l . (1983), as seen i n Table I. Table I Range of Vowel and Vowel Component Durations ( i n msec.) f o r the Present Data and f o r the Strange et a l . (1983) data Present Data Strange et a l . (1983) t o t a l vowel 142 - £48 114 - £0£ ong1i de 15 - 8£ c' c' 30 t a r g e t 45 - 14£ 57 - 1£7 o f f g 1 i de C C 98 33 — 4£ Some of t h i s a d d i t i o n a l v a r i a t i o n was probably due to the m u l t i p l e c o n t e x t s and/or to t r a n s c o n s o n a n t a l c o a r t i c u l a t i o n . O i l of these f a c t o r s c o n t r i b u t e d to making the a p p l i c a t i o n of the c u t p o i n t s l a i d out by Strange et a l . (i983) problematic. In determining how to r e s o l v e these d i f f i c u l t i e s , i n c r e a s e d a t t e n t i o n was paid to the method of determining boundaries between t r a n s i t i o n s and s t e a d y - s t a t e . Most s t u d i e s do not e x p l a i n how e x a c t l y the boundary i s determined, so two methods were compared to see whether the c h o i c e of method would s i g n i f i cant 1y a f f e e t d urat i on values. Each vowe1 was d i v i d e d i n t o t r a n s i t i o n and s t e a d y - s t a t e p o r t i o n s i n two ways, based on s p e c t r o g r a p h i c measurements: (1) the d i v i s i o n point was chosen where a l i n e drawn i n the c e n t e r of the forrnant band departed by more than 1 mm from i t s average p o s i t i o n , and; (£) the d i v i s i o n point was chosen at the i n t e r s e c t i o n of a h o r i z o n t a l l i n e with a l i n e extending the t r a n s i t i o n slope. Table II shows the r e s u l t s obtained with these two methods. These two methods were not s i g n i f i c a n t l y d i f f e r e n t (p<0.05> when measured with a t - t e s t f o r dependent means. The d i f f i c u l t y o f determining the d i v i s i o n between t r a n s i t i o n and s t e a d y - s t a t e i n c r e a s e d as the s l o p e of the t r a n s i t i o n decreased. Tab 1 e II Re 1 a t i v e Pur at ion ( i n "A of t ot a 1 vowel (Jurat i on) of I n i t i a l and F i na1 Trans i t i ons obtained with Two Methods INITIAL FINAL abVba ubVbu ibVbi abVba ubVbu ibVbi V #1 #£ #1 #£ #1 #£ #1 #£ #1 #£ #1 #£ a 19 14 18 £0 31 17 19 14 1£ i ci 19 14 i 31 £1 39 46 £0 16 £3 £8 £7 vie' 18 18 ae £0 £0 16 £0 11 £0 £0 v i C v i e 17 11 E 29 £0 c »-i £7 £ 7 31 40 40 £7 31 4£ 46 I 16 11 £6 £1 15 15 47 48 36 36 vi£> 40 u 13 1 wf £0 1£ a a 38 38 37 £9 53 36 U 16 11 £4 10 9 11 18 16 16 14 9 9 £6 19 £4 £1 1£ a 14 ...iii. 16 19 ! J L 10 e l 37 £9 5£ 6£ 69 59 17 £0 14 7 14 £? oU a a 14 10 11 6 6c 50 £1 18 7£ 57 Methods #1 and #£ are d e s c r i b e d i n the t e x t . The items u n d e r l i n e d are those f o r which both methods r e s u l t i n r e l a t i v e d u r a t i o n s at l e a s t 2.°A l e s s than the r e l a t i v e d u r a t i o n s s p e c i f i e d i n Table I I I . An approach co n s i d e r e d was to t a i l o r - m a k e the e u t p o i n t s to f i t each vowel. T h i s would i n s u r e that no s t e a d y - s t a t e i n f o r m a t i o n was present i n the t r a n s i t i o n segments. T h i s approach was not used f o r s e v e r a l reasons. A major concern was the r e l i a b i l i t y of the measures of the n e a r - h o r i z o n t a l t r a n s i t i o n s . In s e v e r a l case, F£ was h o r i z o n t a l , and movement measures were based on F l or F"3a T h i s would mean that the major cuts, F£, was a l r e a d y at i t s t a r g e t . Pis w e l l , the b r e v i t y of the d u r a t i o n s of these t r a n s i t i o n s was such that l e s s than £0"/. of the t o t a l s t i m u l u s would be presented i n some cases, thus p o s s i b l y making the task too d i f f i c u l t . F'inally, i t was f e l t that s i n c e the e f f e c t s of the subgroup of vowels </U, u, ou, "'/) co u l d be separated and examined afterwards, using d i f f e r e n t cut p o i n t s f o r each vowel was unriecessary. Two changes were made though. F i r s t l y , s i n c e the r a t i o n a l e f o r the in t e r m e d i a t e grouping (/i,u/) used i n Strange et a l . ( 1 3 8 3 ) was-; not apparent, i t s members were maintained as an in t e r m e d i a t e group f o r the sake of comparison but a l s o i n c l u d e d i n t h e i r c o r r e c t grouping (based on L e h i s t e & Peterson, 1 9 6 1 ) , with the other long simple n u c l e i vowels, r e s u l t i n g i n a double grouping f o r them. Secondly, the long complex n u c l e i were t r e a t e d as a separ a t e group. The d i v i s i o n s f o r /el'/ had a 30"/. o n g l i d e p o r t i o n and a 10"/. o f f g l i d e p o r t i o n , while f o r /oU/, i t was the reverses 10"/. f o r the o n g l i d e and 30"/. f o r the o f f g l i d e . These percentages were chosen with r e f e r e n c e t o the value s obtained from L e h i s t e and Peterson ( 1 9 6 1 ) as well as t o those obtained from the present speaker's vowels ac r o s s contexts. In summary then, the cut p o i n t s that were chosen were based on L e h i s t e and Peterson ( 1 9 6 1 ) , Strange et. a l ( 1 9 8 3 ) , and the present speech m a t e r i a l s . The vowels were separated i n t o four groups; short n u c l e i (/U, E, I /) , i n t e r m e d i a t e n u c l e i ( / i , u / ) , long simple n u c l e i (/a,ae, i , u/) , and long complex n u c l e i %i 6 (/eI,oUY). The percentages used are d e t a i l e d i n Table I I I . These percentages are not p e r f e c t matches f o r the a c t u a l r e l a t i v e d u r a t i o n s of each vowel st i m u l u s . The mismatches that r e s u l t i n some s t e a d y - s t a t e i n f o r m a t i o n being i n c l u d e d i n the t r a n s i t i o n p o r t i o n s are u n d e r l i n e d i n Table II. Table III R e l a t i v e Durations ( i n "/» of t o t a l vowel d u r a t i o n )  of I n i t i a l and F i n a l T r a n s i t i o n s Chosen f o r Each Vowel VOWEL GROUP MEMBERS INITIAL FINAL short /U/,/•-•/, / E / , / I / 15 35 " i n t e r m e d i a t e " / i / , / u / 15 £5 long simple /a/,/ae/, / i / , / u / 15 £0 long complex / e l / 30 10 long complex /oU/ 10 30 The groups and d u r a t i o n s shown here, other than the i n t e r m e d i a t e group, are based on d e s c r i p t i o n s by L e h i s t e and Peterson (1961). The i n t e r m e d i a t e group i s based on Strange et. a l (1983). The V i C V e C V i ' s were reduced to -CV eC—'s and each vowel (V B) was then cut, a c c o r d i n g to i t s category. The s t i m u l i f o r the C o n t r o l c o n d i t i o n (CO) remained u n a l t e r e d a f t e r being reduced to /bVb/» The s t i m u l i f o r the Steady-State Only c o n d i t i o n (SS) was composed of o n l y the c e n t e r p o r t i o n of the vowel. The s t i m u l i f o r the T r a n s i t i o n s Only c o n d i t i o n (TR) were made up of the i n i t i a l and f i n a l t r a n s i t i o n s , separated by 37 a s i l e n t i n t e r v a l , whose d u r a t i o n was equal to the d u r a t i o n of the absent c e n t e r p o r t i o n . The composition of the s t i m u l i were such that any t i m i n g cues a v a i l a b l e from the r e l a t i v e t r a n s i t i o n and t a r g e t d u r a t i o n s were s t i l l present a l l the ex per i merit a I cond i t i ons. The r e l a t i v e i n t e n s i t y of the vowels, as measured on the c o n t r o l -CV eC-'s, was then compared to the v a l u e s found i n L e h i s t e and Peterson (1959). S e v e r a l recorded vowels showed s i g n i f i c a n t l y a t y p i c a l amplitudes, p o s s i b l y as a r e s u l t of the speaker's e f f o r t s t o maintain d i s t i n c t i o n s i n the face of the a r t i c u l a t o r y c o n s t r a i n t s that had to be met. Those vowels not f a l l i n g i n t o a high, rued i urn, 1 ow i n t e n s i t y c 1 a s s i f i c a t ion a c c o r d i n g to L e h i s t e and Peterson (1959) were adju s t e d by r e p l a c i n g the o r i g i n a l amplitude with the group mean, thus a c h i e v i n g a b e t t e r agreement with L e h i s t e and Peterson's data, as seen i n Table IV. Vowe 1 Tatal I nt e r i s i t i es ( i n dB) e IV a f t e r adj ust merit accord i no to L e h i s t e and Peterson (1959) IMTEMSITY V CONTEXT aCVCa uCVCu iCVCi High a 77. 1 77. 1 79. 1 oU 78. £ 77. 6 (75.8) 79. 6 77.6 (80 .13)* 77. 1 77. 6 ae 78. 7 77. 6 77. 6 E 76. 74. 1 77. 1 (80.0) Med. e l 75. 8 74. 6 74. 1 U 74. 6 76. 5 77. 6 u 74. 6 71. 6 7£. 6 I 73. 6 74. 6 (78.7) 74. 1 (80.0) Low i 71. 6 71.6 (73.6) /£. 6 The v a l u e s i n parentheses are the o r i g i n a l i n t e n s i t i e s o f the vowels not f i t t i n g i n t o the high, medium, 1 ow d i s t r i but i on of Peterson and L e h i s t e (1959) l i s t e d i n the r e g u l a r columns. T h e i r adjusted v a l u e s are The e d i t e d s t i m u l i were lowpass f i l t e r e d at £.5 kHz and recorded onto audiotapes on a r e e l — t o — r e e l Revox A77 tape r e c o r d e r f o r p r e s e n t a t i o n t o the s u b j e c t s . 4.£ Experimental C o n d i t i o n s There were th r e e experimental c o n d i t i o n s , corresponding t o the t h r e e s e t s of s t i m u l i d e s c r i b e d above: 1) C o n t r o l (CO); £) Steady-State Only (SS) ; 3) T r a n s i t i o n Only (TR) . Within each c o n d i t i o n , each of the ten vowels occurred i n th r e e contexts, r e s u l t i n g i n a t o t a l of 30 s t i m u l u s types. In a d d i t i o n , f o r the second and t h i r d c o n d i t i o n s , two of the vowels (/i,u/) were mod i f i e d i n two ways (see Table I I I ) , r e s u l t i n g i n twelve vowel types and three contexts, f o r a t o t a l of 36 s t i m u l u s types. Each s t i m u l u s type was repeated f i v e times f o r each c o n d i t i o n . Each c o n d i t i o n was preceded by f i v e b u f f e r items which were not scored. T h i s r e s u l t e d i n a t o t a l of 155 items f o r the C o n t r o l C o n d i t i o n (CO) and 185 items f o r the Steady-S t a t e Only (SS) and the T r a n s i t i o n s Only (TR) c o n d i t i o n s . For each c o n d i t i o n , p r e s e n t a t i o n order of the s t i m u l i was randomized with the c o n s t r a i n t that each vowel had to occur once i n each block of 30 (or 36 items, depending on c o n d i t i o n ) , and no vowel (V B) could occur twice c o n s e c u t i v e l y . 4. 3 S u b j e c t s L i s t e n e r s were £0 graduate students, 13 women and one man, from the speech s c i e n c e department. They were not paid f o r t h e i r t i me. Pr i or t o the ex per i ment, each was q uest i oned on h i s or her language background and hearing a b i l i t y . fill p a r t i c i p a n t s (speaker, t r a n s c r i b e r s , and l i s t e n e r s ) had some phonetics t r a i n i n g and were n a t i v e speakers of (standard) Canadian E n g l i s h , most being from r e g i o n s of Western Canada. 40 4.4 Procedure Pill s u b j e c t s were administered a l l the c o n d i t i o n s . Tapes were presented to the s u b j e c t s i n d i v i d u a l l y over Beyer DT48 headphones i n an IPlC soundproof booth, at 50-60 dB SPL. The s u b j e c t s were t o l d that the s t i m u l i were e l e c t r o n i c a l l y m odified vowels. They were asked t o i d e n t i f y the vowels heard by c i r c l i n g the a p p r o p r i a t e keyword. The keywords weres beeb, bib, babe, beb, bab, bob, bub, bobe, buub, boob. Pi l i s t of corresponding phonetic symbols occurred a f t e r every ten items on the answer sheet t o a i d the s u b j e c t s i n i n t e r p r e t i n g the o r t h o g r a p h i c r e p r e s e n t a t i o n s . The ten c h o i c e s were reviewed once before beginning t e s t i n g . PI copy of the i n s t r u c t i o n s t o the s u b j e c t can be found i n the appendix. The t e s t i n g period, c o n s i s t i n g of a p r a c t i c e p e r i o d and t h r e e t e s t periods, with short breaks between, was approximately one and a h a l f hours long. The p r a c t i c e items were always presented f i r s t . P r e s e n t a t i o n order f o r the t e s t c o n d i t i o n s was balanced a c r o s s s u b j e c t s . The ten item p r a c t i c e p e r i o d was composed of a subset of the c o n t r o l s t i m u l i . Each vowel type occurred once, but the c h o i c e of context was random. Feedback was not provided. Items were presented at an ISI of 4 seconds, with an 8 second gap o c c u r r i n g a f t e r every block of ten. Pi 1000-H:-: beep was i n s e r t e d i n the c e n t e r of the gap t o help s u b j e c t s keep t h e i r p l a c e on the answer sheets, s i n c e the s t i m u l i were not preceded by any spoken numbers. 41 Chapter S RESULTS 5.1 O r g a n i z a t i o n of the Data A de-randomization program was w r i t t e n t o s o r t the s u b j e c t s ' responses. A f t e r the data was entered and de-randomized, i t was c o l l a p s e d i n t o a number of c o n f u s i o n m a t r i c e s to allow examination of the p a t t e r n of responses, using s u b j e c t , context and c o n d i t i o n as the independent v a r i a b l e s . The matrices were i n i t i a l l y o rganized as 10x10 g r i d s (ten s t i m u l i by ten responses (10x10) f o r the CO c o n d i t i o n , and 12x10 f o r the SS and TR c o n d i t i o n s . The .1.2x10 ma t r i c e s were a c t u a l l y the r e s u l t of c o l l a p s i n g two 10x10 matrices, each i d e n t i c a l except f o r the presence of a d i f f e r e n t v e r s i o n ( i n t e r m e d i a t e or long grouping) of / i / and /u/. Because the two v e r s i o n s of / i / and /u/ were found t o y i e l d i d e n t i c a l r e s u l t s , the i n t e r m e d i a t e grouping of Strange et a l . (1983) was e l i m i n a t e d , r e s u l t i n g i n 10x10 m a t r i c e s f o r the SS and TR c o n d i t i o n s as w e l l as f o r the CO c o n d i t i o n . The 10x10 mat r i c e s w i l l be r e f e r r e d t o as the unmodified matrices i n d e s c r i p t i o n s t o f o l l o w . In a d d i t i o n , f o r reasons t o be s t a t e d i n f o l l o w i n g s e c t i o n s , /a/ and /'"/ r e s u l t s were c o l l a p s e d , and data r e g a r d i n g /U/ were e l i m i n a t e d . The r e s u l t i n g m a t r i c e s are r e f e r r e d t o as the modified matrices, and are considered t o present a c l e a r e r summary of the s u b j e c t s ' p e r c e p t i o n s than the 10x10 matrices. ANOVAs and Newman-Keu1s were then performed on the t o t a l c o r r e c t responses f o r both the unmodified and the modified m a t r i c e s t o determine any s i g n i f i c a n t e f f e e t ( s ) due t o c o n d i t i o n , context, i n t e r a c t i o n or s u b j e c t . Post-hoc t r a n s c r i p t i o n s ( d e s c r i b e d i n the d i s c u s s i o n ) by four of the s u b j e c t s were performed l a t e r i n an attempt t o e x p l a i n some of the e r r o r p a t t e r n s encountered. The f o l l o w i n g r e s u l t s and d i s c u s s i o n s e c t i o n s i n v o l v e e r r o r r a t e s r a t h e r than c o r r e c t performance r a t e s . E r r o r r a t e s were used because the s t u d i e s on which a s p e c t s of t h i s experiment are based (Strange et a l . , 1983; Jen k i n s et a l . 1983) r e p o r t e d t h e i r f i n d i n g s i n terms of e r r o r r a t e s . In the c o n f u s i o n m a t r i c e s of Table V, c o r r e c t responses are l o c a t e d on the d i a g o n a l , and e r r o r s are responses o c c u r r i n g anywhere o u t s i d e o f the d i a g o n a l . 5.2 Performance over C o n d i t i o n s and Contexts Ta b 1 e V shows the nine unrnod i f i ed con f us i on mat r i ces f o r c o n d i t i o n and context, each summated over a l l the s u b j e c t s . The t o t a l c o r r e c t responses, c a l c u l a t e d by summing the responses from the d i a g o n a l s of the c o n f u s i o n m a t r i c e s i n Table V were s u b t r a c t e d from the t o t a l p o s s i b l e . The va l u e s remaining are the e r r o r r a t e s and are shown f o r each c o n d i t i o n and context i n Table VI. Table V Summated Confusion Matrices for All Subjects (la) CO - /a/ resp i I el E aE a A oU U u stim i 79 13 8 I - 83 - 1 7 - - - - - -el - - 96 4 - - - - - -E - 1 - 97 2 - - - - -aE - - - - 74 15 11 - - -a 3 97 - - -A _ _ _ _ _ 3 6 6 3 _ 1 _ oU - - - - - - - loo - -U _ _ _ _ _ _ _ _ 1 0 o -u 2 98 (2a) SS - /a/ resp i I el E aE a A oU U u stim i i 81 10 9 - - - - - - -io 79 19 2 - - - - - - -I - 9 1 - 9 - - - - - -el l 17 72 10 - - - - - -E _ - - i o o - - - - - -aE - - - - 90 8 2 - - -a - - - - 1 14 85 - - -A 42 5 8 - - -oU ioo - -U _ _ _ _ _ _ 6 _ 9 4 _ ui - - - - - - - 1 16 83 u 2 - - - - - - - 2 8 90 (3a) TR - /a/ resp i I e l E a E a / \ o U U u stim 11 4 9 85 - - - - - - -1 2 4 13 82 - - - - - - -I - 1 3 - 8 7 - - - - - -el - 2 72 25 - - - - - -E - - - 50 49 1 - - - -aE - - - 2 56 17 25 - - -a - - - - - 6 92 - - -A - - - - - 48 51 1 - -oU _ _ _ _ _ _ _ 1 0 0 _ _ U - _ _ _ _ _ 2 4 94 -ui - - - - - - 1 3 32 64 u 2 - - - - - - - 1 29 70 (lb) CO - /u/ resp i I el E ae a A oU U u stim i 98 2 - - - - - - - -I - 9 9 - i _ _ _ _ _ _ el - _ 9 9 _ i _ _ _ _ _ E - _ - i o o - - - - - -aE - - - - 100 - - - - -a - - - - - 22 74 - - -A - - - - - 39 61 - 4 -oU _ _ _ _ _ _ _ ioo - -U 5 90 - 5 -u 5 95 (2b) SS - /u/ resp i I el E ae a' A oU U u stim 11 98 1 1 - - - - - - -12 100 - - - - - - - - - -I - 9 8 - 2 - - - - - -el - - ioo - - - - - - -E - - - 100 - - - - - -aE - _ i - 97 2 _ _ _ _ a - - - - 1 21 77 - - -A - - - - 2 37 61 - - -oU ioo - -U 3 94 - 3 -ui - - - - - - - 1 4 95 u 2 - - - - - - - 1 1 98 (3b) TR - /u/ resp i I el E ae a A oU TJ u stim 11 66 7 27 - - - - - - -1 2 60 13 26 1 - - - - - -I - 70 - 30 - - - - - -el - - 99 1 - - - - - -E - - _ 77 23 _ _ _ _ _ aE - - - 5 89 3 3 - - -a - - - - 2 14 84 - - -A - - - - - 38 61 - 1 -oU _ _ _ _ _ _ _ gl _ _ U - - - - - 20 77 - 3 -ui - - - - - - - - - 100 u 2 99 U4 Table V (cont.'d) (lc) CO - / i / (2c) SS - / i / (3c) TR - / i / resp i I el E ae a A oU u u resp i I el E ae a A oU U u resp i I el E ae a A oU U u stim stim stim i 100 i l 98 2 i l 90 1 9 I 2 97 - 1 i-2 100 i2 86 - 13 - - - 1 - - -el 1 - 99 I - 100 I - 95 - 5 ' - - - - - -E - - - 100 - - - - - - el 24 - 76 el - - 98 2 aE - - - - 100 - - - - - E - - - 100 - - - - - - E - - - 84 15 1 - - - -a 89 9 2 - - aE - - - - 99 1 - - - - aE - - - 2 97 1 - - - -A 9 91 - - - a 98 2 - - - a 66 34 - - -oU - - - - - - - 100 - - A 10 90 - 1 - A - - - - 1 22 77 - - -U 100 - OU 99 - 1 OU 99 - 1 u 1 99 u "I U2 1 9 1 3 2 90 96 98 u u l U2 1 6 93 2 6 98 94 (Id) CO - 3 contexts resp i I el E ae a A oU U u stim i 277 15 8 - - - - - - -I 2 279 - 19 - - - - - -el i _ 294 - - - - - - -E - 1 - 297 2 aE - - - 1 273 15 11 - - -a - - - - - 114 180 2 4 -A - - - - - 84 215 - 1 -oU _ _ _ _ _ _ _ 299 - -U - - - - - 5 90 - 205 -u - - - - - - - - 8 292 (2d) SS - 3 contexts resp i I el E ae a A oU U u stim i , 277 13 10 - - - - - - -i2 279 19 2 - - - - - - -I - 289 - 11 - - - - - -el 25 17 248 10 - - - - - -E _ _ _ 300 - - - - - -aE - - 1 - 286 11 2 - - -a - - - - 2 133 164 - - -A - - - - 2 89 209 - - -oU _ _ _ _ _ _ _ 299 - -U - - - 1 - 3 109 - 187 -ui - - - - - - 1 5 20 274 u, - - - - - - - 5 9 286 (3d) TR - 3 contexts resp i I el E ae a A oU U u stim i-l 160 17 121 2 - - - - - -i 2 150 26 121 2 - - 1 - - -I - 178 - 122 - - - - - -el - 2 269 27 - - - - - -E - - - 211 87 2 - - - -aE 9 242 21 28 - - -a - - - - 2 86 210 - 1 -A - - - - 1 108 189 1 1 -oU _ _ _ _ _ _ _ 290 - 10 U - - 1 - 20 85 4 190 ai - - - - - - 1 3 34 262 - - - - 1 - - 1 35 263 The summated confusion matrices for the twenty subjects are printed above. The matrices are coded by condition and context. The SS and TR conditions are asymmetrical as a result of collapsing the 10x10 confusion matrices occurring for each of the two groupings of / i / and /u/ (#1 = intermediate and #2 = long) into 10x12 matrices for each condition and context combination. 4^  45 T a b l e VI L r r o r R ate ( i n "/) a c r o s s C o n d i t i o n s and C o n t e x t s CONTEXT CONDITION / a / /u/ / i / CO SS TR £0. 4 £1. £ 48. 4 The v a l u e s f r o m T a b l e VI show e r r o r r a t e s a s h i g h a s £0°/ f o r t h e CO c o n d i t i o n and o v e r 33% f o r two o f t h e t h r e e c o n t e x t s i n t h e TR c o n d i t i o n . T h i s i s a c t u a l l y m i s l e a d i n g , b e c a u s e , on e x a m i n a t i o n o f t h e c o n f u s i o n m a t r i c e s o f t h e c o n t r o l c o n d i t i o n , i t i s a p p a r e n t t h a t t h e low s c o r e s a r e p r i m a r i l y c a u s e d by two f a c t o r s ; (1) c o n f u s i o n o f / a / and / '/ w i t h e a c h o t h e r (£64/£71 o r 97°/ o f t h e t o t a l m i s c l a s s i f i c a t i o n s f o r b o t h , from mat r i c e (Id) o f T a b l e V) and 5 (£) m i s c l a s s i f i c a t i o n o f /U/ i n t h e /u/ c o n t e x t a s / a / o r /"-/ (95/100 o r 95"/. o f t h e t o t a l p r e s e n t a t i o n t i m e s f o r /U/, from m a t r i c e ( l b ) o f T a b l e V ) . The c o n f u s i o n o f / a / and /'"/ a p p e a r s t o r e s u l t from a s i m i l a r i t y i n p r o d u c t i o n , a r i s i n g from t h e l a c k o f c o n t r a s t i v i t y i n t h e d i a l e c t o f s p e a k e r and s u b j e c t s , s i n c e t h e c o n f u s i o n was c o n s i s t e n t o v e r s u b j e c t s , few o t h e r c o n f u s i o n s o c c u r r e d , and s p o n t a n e o u s s u b j e c t i v e j u d g e m e n t s o f d i f f i c u l t y i n d i f f e r e n t i a t i n g / a / and /'"/ were r e p o r t e d . Pissman e t a l . 46 (1982), who drew s u b j e c t s from the same d i a l e c t a l r e g i o n as those i n the present study, used the / " - / - / O / c o n t r a s t i n s t e a d . The extremely r e g u l a r m i s c l a s s i f i c a t i o n of the s t i m u l u s /UY i n the /u/ context as an /a/ or an /'"/ suggests that t h i s vowel type was p o o r l y produced, although i t had been t r a n s c r i b e d c o r r e c t l y when presented i n i t s o r i g i n a l V j C V a C V j form. Based on these two r a t i o n a l e s , c e r t a i n adjustments to the m a t r i c e s were made. The /a/ and /•'"/ s t i m u l i were c o l l a p s e d i n t o a s i n g l e s t i m u l u s type. The number of combined responses was d i v i d e d by two to ensure e q u i v a l e n t weighting compared to the other vowels. fill c o n f u s i o n s , then, of /a/ and /•'•/ with each other were counted as c o r r e c t , and a l l other e r r o r s were maintained (with the a b s o l u t e s c o r e s f o r both c o r r e c t and i n c o r r e c t h a l v e d ) . Pin example of a r e s u l t of t h i s m a n i p u l a t i o n i s s c o r e of 98 c o r r e c t and £ i n c o r r e c t obtained f o r /&/ + /'"/ i n the matrice (lb) of Table V . Next, the vowel type /U/ was e l i m i n a t e d from the a n a l y s i s , due to the low performance on t h i s vowel type i n the /u/ context, even i n the c o n t r o l c o n d i t i o n . Although performance i n the other two c o n t e x t s was good, /UY was a l s o e l i m i n a t e d f o r those c o n t e x t s . T h i s was done to keep the c o n t e x t s e q u i v a l e n t i n terms of the number of vowel c h o i c e s thereby reducing the number of s t a t i s t i c a l asymmetries t o be d e a l t with. The v a l u e s remaining y i e l d e d the modified t o t a l s . The sum of these t o t a 1 s f o r a l l s u b j e c t s f o r each comb inat ion of 47 context and c o n d i t i o n was s u b t r a c t e d from the t o t a l responses p o s s i b l e . The r e s u l t i n g e r r o r r a t e s are shown as percentages i n Table VII. They are con s i d e r e d t o present a c l e a r e r summary of the performance i n each of the c o n d i t i o n s and c o n t e x t s than the unmodified m a t r i c e s and t h e i r corresponding s c o r e s and e r r o r r a t e s . Table VII Mod i f ied E r r o r Rates ( i n "A) a c r o s s C o n d i t i o n s and Contexts CONDITION /a/ CONTEXT /u/ / i / C o n t r o l Steady-State T r a n s i t ions 10. £ 41.6 1. 5 1. 9 14. 8 1. 0 -J cr 5 . £ The modified v a l u e s show high performance ( l e s s than 10% e r r o r s ) i n the CO c o n d i t i o n f o r a l l c o n t e x t s : the e r r o r r a t e i s 9. £"/. f o r the /a/ context, as compared to l e s s than 1.5°/. f o r the othe r two contexts. Performance i s s i m i l a r f o r the SS c o n d i t i o n , with 10. £"/., 1.9%, and 3. 5°/. e r r o r s i n /a/, /u/ and / i / c o n texts, r e s p e c t i v e l y . The TR c o n d i t i o n shows a marked drop i n one of the contexts, with e r r o r r a t e s of 41.6% f o r the /a/ context, as compared t o 14.8"/. f o r the /u/ context and 5. £"/. 48 f o r the / i / c ontext. The s t a t i s t i c a l a n a l y s e s were performed on both the modified and unmodified t o t a l s f o r comparison. The ANOVA and the Neuman-Keuls t e s t s had s i m i l a r r e s u l t s (concerning s i g n i f i c a n c e ) f o r c o n d i t i o n and i n t e r a c t i o n e f f e c t s . T h i s should be expected s i n c e the d i f f i c u l t y with /a/ and /'"/ occurred a c r o s s a l l the co n t e x t s and c o n d i t i o n s , r e s u l t i n g i n g e n e r a l l y depressed performance scores. The major d i f f e r e n c e between the unmodified and the modified data arose i n comparisons of performance a c r o s s c o n t e x t s ; i t was caused p r i m a r i l y by the app^irent s e l e c t i v e d i f f i c u l t y i n p e r c e i v i n g the /U/ s t i m u l i . S i n c e the d i f f i c u l t y o r i g i n a t e d most c e r t a i n l y i n the production, not the p e r c e p t i o n of /U/, the r e s u l t s f o r the modified t o t a l s , where /UV i s not include d , w i l l be the ones d i s c u s s e d i n d e t a i l . A treatment-by-treatment-by-subject ANOVA was performed on the t o t a l c o r r e c t responses f o r each c o n d i t i o n and context. Th e ANOVA sh owed a h i g h l y s i g n i f i cant cond i t i on e f f e e t , a h i g h l y s i g n i f i c a n t context e f f e c t , and a h i g h l y s i g n i f i c a n t i n t e r a c t i o n e f f e c t with F va l u e s of 115.42, 107.64 and 49.51 r e s p e c t i v e l y . The r e s u l t s are d i s p l a y e d i n Table VIII. Table VIII ANOVA Re s u l t s f o r both Unmodi f ied T o t a l s and the Mo d i f i e d T o t a l s UNMODIFIED TOTALS SOURCE SS df rns F p T o t a l 9433.£0 179 Su b j e c t s 376.76 19 - ' -Treatment 1 (Conditions) £408.13 2 1204.07 85.85 p<.01 Treatment 2 (Contexts) 4669.20 2 2334.60 323.88 p<.01 Tr. 1 x Tr. 2 761.47 4 190.37 35.22 p(.01 E r r o r Treatment 1 532.98 38 14.03 - -E r r o r Treatment £ 273.91 38 7.21 -E r r o r Tr. 1 x Tr. £ 410.76 76 5. 40 -MODIFIED TOTALS SOURCE SS df ms F p T o t a l 5£70.95 179 _ _ Subject £38.95 19 Treatment 1 (Conditions) 17.23.60 £ 861.80 115.42 p<.01 Treatment £ (Contexts) 1565.63 £ 78£. 8£ 107.64 p<.01 Tr. 1 x Tr. £ 854.67 4 213.67 49.51 p<.01 E r r o r Treatment 1 £83.73 38 7.47 — E r r o r Treatment £ £76.37 38 7.£7 -E r r o r Tr. 1 x Tr. £ 3£8.00 76 4. 3£ — — A Neuman-Keuls a n a l y s i s was used to examine which p a i r s were causing the s i g n i f i c a n t d i f f e r e n c e s . The p a i r - w i s e comparison showed that the c o n d i t i o n e f f e c t , with context c o l l a p s e d , was caused by s i g n i f i c a n t d i f f e r e n c e s 50 TR and the SS c o n d i t i o n s at p < -05 (degrees of freedom (df) beginning at 59 and 3). The CO and the SS c o n d i t i o n s were not s i g n i f i c a n t l y d i f f e r e n t up to p < .05. The context e f f e c t , with c o n d i t i o n s c o l l a p s e d , was found to be s i g n i f i c a n t f o r /a/ versus /u/ and /a/ versus / i / , a gain at p < .05, (df beginning at 59 and 3). The c o n t e x t s /u/ and / i / were not s i g n i f i c a n t l y d i f f e r e n t . The i n t e r a c t i o n s were a l s o t e s t e d with the Neuman-Keuls (df beginning at 19 and 3). Performance f o r the TR c o n d i t i o n -/a/ context combination was found t o be s i g n i f i c a n t l y d i f f e r e n t from every other t o t a l at p < .01. None of the other comparisons, i n c l u d i n g those with the /a/ context i n the other c o n d i t i o n s , were s i g n i f i c a n t at p < .05. From these analyses, i t can be seen that TR s t i m u l i are i d e n t i f i e d s i g n i f i c a n t l y more p o o r l y than SS or CO s t i m u l i . The /a/ context was found t o be a s i g n i f i c a n t detriment to vowel i d e n t i f i c a t i o n as compared to the / i / and /u/ contexts. The combination of c o n d i t i o n and context r e s u l t e d i n s i g n i f i c a n t l y poorer performance i n the /a/ context f o r the TR c o n d i t i o n as compared to every other combination of context and cond i t i on. The /a/ context y i e l d e d the lowest s c o r e i n every c o n d i t i o n , producing a s i g n i f i c a n t context e f f e c t , but not producing s i g n i f i c a n t i n t e r a c t i o n comparisons except i n the TR c o n d i t i o n . Based on t h i s , an a d d i t i o n a l Newman-Keuls was performed t o compare c o n d i t i o n s , without the i n f l u e n c e o f the /a/ context on the scores. Without the /a/ context, the TR c o n d i t i o n was not found t o be s i g n i f i c a n t l y d i f f e r e n t from the SS or CO c o n d i t i o n s up to p < .85 (df beginning at 39 and 3). T h i s i n d i c a t e s that the TR s t i m u l i are perc e i v e d as well as the SS or the CO s t i m u l i f o r the /u/ and / i / c o ntexts. The r e s u l t s not only support the s u f f i c i e n c y of t r a n s i t i o n s as cues t o vowel i d e n t i f i c a t i o n , but show, f o r two out of t h r e e contexts, performance l e v e l on t r a n s i t i o n s only s t i m u l i t o be at a l e v e l comparable t o that of c o n t r o l and s t e a d y - s t a t e only s t i m u l i . ft marked context e f f e c t , however, q u a l i f i e s t h i s c o n c l u s i o n by depressing performance i n the TR c o n d i t i o n f o r the context /a/. 5.3 Comparison of the two medial / i / and /u/ groupings Two d i f f e r e n t e u t p o i n t s had been used f o r the / i / and /u/ vowel n u c l e i : one conforming t o the Strange et a l . (19S3) and the J e n k i n s et a l . (19S3) i n t e r m e d i a t e grouping, and one p l a c i n g these two vowels i n L e h i s t e and Peterson's (1961) long simple n u c l e i grouping. Performance was very s i m i l a r f o r the two v e r s i o n s of / i / and /u/ i n each c o n d i t i o n (see Table IX). Comparison of performance on these items with a t - t e s t f o r dependent means found no s i g n i f i c a n t d i f f e r e n c e s at p < .05 f o r e i t h e r the SS or the TR c o n d i t i o n s . Table IX Comparison of E r r o r Rates ( i n %) f o r Two Groupings of / i / and /u/ SS cond i t i on TR c o n d i t i o n VOWEL intermed. long intermed. long / i / /u/ 7. 7 8. 7 7. 0 8. 0 46. 7 1£. 7 Eiased on these f i n d i n g s , the s t i m u l i from the in t e r m e d i a t e grouping were dropped from f u r t h e r a n a l y s i s . 5.4 Item A n a l y s i s The e r r o r d i s t r i b u t i o n was not uniform over the ten vowel types as can be seen i n the c o n f u s i o n m a t r i c e s of Table V,. E r r o r s tended t o be one to two vowels away around the vowel "loop", when p l o t t e d i n an Fl/F£ graph, with c o n f u s i o n s o c c u r r i n g f o r s p e c t r a l l y s i m i l a r members. E r r o r s were not l i m i t e d t o c o n f u s i o n s w i t h i n a s i n g l e vowel category (e.g. tense vowels). E r r o r r a t e s f o r each vowel type are d i s p l a y e d i n Table X. Table X E r r o r Rates ( i n "/•) as a F uncti o n of Vowel Type f o r A l l Co nd i t i on s and Contexts CO co nd i t i on V /a/ /u/ / i / i £1 C 0 I 17 1 e l 4 1 1 E 0 0 ae £6 1 0 a 97 (1) 78 (£) 11(1) 37 39. 9 oU 0 0 1 U 0 95 0 u C 5 1 TR c o n d i t i o n SS c o n d i t i o n V /a/ /u/ / i / V /a/ /u/ / i / i 96 40 14 i £1 0 0 I a? 30 5 I 9 £ 0 e l £8 1 C e l £8 0 £4 E 50 £3 16 E 0 0 0 ae 44 11 3 ae 10 3 1 a 94(£) 86(£) %3 \2r ( .1 ) a 86(1) 79(£) £(0) 49 39 4£ 39 10 oU 0 9 1 oU 0 0 1 U 6 97 7 U 6 97 10 u 30 1 6 u 10 £ £ The numbers i n parentheses the e r r o r r a t e s a f t e r c o l l a p s i n g arid / '"• / i nt o a s i n g l e vowe I type. The o r i g i n a l s c o r e s are the reg u1ar co1umns, The e r r o r p a t t e r n v a r i e d somewhat over c o n d i t i o n s and contexts. C o n s i s t e n t l y good performance was obtained only f o r the /oU/ and /u/ vowels. For the TR c o n d i t i o n — / a / context com b i nat i on, wh ere the maj or i t y of e r r o r s occurred, the high est e r r o r r a t e s were f o r / i / (96"/.), / I / (87°/.), and /E/ (50%). Of these e r r o r s , the m a j o r i t y of c o n f u s i o n s were / i / to / e i / (82%, with 14"/ other e r r o r s and 4"/ c o r r e c t ) , / I / to /E/ (87"/, with 13"/ c o r r e c t ) , and /E/ to /ae/ (49°/, with 1% other e r r o r s and 50"/ c o r r e c t ) . These e r r o r s , as well as those f o r the other vowels, are q u i t e c o n s i s t e n t , suggesting that the e l i m i n a t i o n of s t e a d y - s t a t e i n f o r m a t i o n does not r e s u l t i n ambiguous s t i m u l i , but i n c a t e g o r i c a l changes i n vowel q u a l i t y , f o r example, / i / i n the /a/ context becomes / e i / upon removal of the st eady—st at e port i on of the vowe? 1. Chapter 6 DISCUSSION 6.1 Vowel Cues A v a i l a b l e i n TR S t i m u l i Performance i n two of the th r e e c o n t e x t s of the TR c o n d i t i o n was as good as performance i n the CO and SS c o n d i t i o n s . I f good performance r e s u l t e d i n the TR c o n d i t i o n , i t was o r i g i n a l l y hypothesized t o be due to the s u f f i c i e n c y of t r a n s i t i o n i n f o r m a t i o n as cues t o vowel i d e n t i t y . Examination of other p o s s i b l e reasons should occur before c o n c l u d i n g that high performance on s t i m u l i without s t e a d y - s t a t e s i s a c t u a l l y due t o the presence of the t r a n s i t i o n s . Sources of i n f o r m a t i o n s t i l l remaining i n the e d i t e d s t i m u l i other than t r a n s i t i o n s c ould play a s i g n i f i c a n t r o l e i n the i d e n t i f i c a t i o n process. Th i s i nf ormat i on i nc 1 udes d urat i on, d i pht hong i zat i on ( forrnant movement), and i nt e n s i t y d i f f e r e n c e s . I f d u r a t i o n i s an important cue, then i t s maintained presence should a i d d i f f e r e n t i a t i o n of two groups of vowelss long (tense) and short ( l a x ) . Examination of the TR matrix (3d) of Table V r e v e a l s that the c o n f u s i o n s of the vowel types with the highest e r r o r r a t e s , namely / i / t o / e i / and / I / to /E/, are both w i t h i n the short vowel group, or w i t h i n the long vowel group, but matrix (3d) a l s o shows that many of the other c o n f u s i o n s are between these two groups. D u r a t i o n may be a u s e f u l cue, but i t s presence here does not prevent t e n s e - l a x 56 c o n f u s i o n s . T h i s f i n d i n g i s f u r t h e r supported by the f i n d i n g of Strange et a i . (1983) that p a r t i a l n e u t r a l i z a t i o n of d u r a t i o n does not s i g n i f i c a n t l y a f f e c t p e r c e p t i o n of vowels with the s t e a d y — s t a t e removed. Forrnant movement r e s u l t i n g from d i pht hong i zat i on i s d i f f i c u l t t o separate from that r e s u l t i n g from consonant-vowel t r a n s i t i o n s . The obvious diphthongs (/ai, o i , all/) , d e s c r i b e d o r i g i n a l l y by L e h i s t e and Peterson (1961), were not i n c l u d e d i n the present vowel set. G l i d e d vowels (/ei,o(J/), however, were used. For the these vowels, the g l i d e and the consonant-vowel t r a n s i t i o n were not separated? the e u t p o i n t s were adjus t e d to encompass the e n t i r e p e r i o d of movement. As a r e s u l t , these s t i m u l i had the a d d i t i o n a l i n i t i a l or f i n a l g l i d i n g cue a v a i l a b l e . Examination of performance (see matrice (3a), Table V) shows that the vowels / e i / and /oU/ had among the best i d e n t i f i c a t i o n s c o r e s i n the c r i t i c a l combination (where o v e r a l l performance was worst); TR c o n d i t i o n — / a / context. The g l i d i n g i n f o r m a t i o n present i n the t r a n s i t i o n segments may, t h e r e f o r e , have been u s e f u l i n i d e n t i f i c a t i o n of the TR st i r n u l i . Normal i n t e n s i t y cues were maintained and even provided wherever the t y p i c a l r e l a t i v e i n t e n s i t y d i d not occur at p r o d u c t i o n of the vowels. T h i s i n f o r m a t i o n i s present i n a l l the c o n d i t i o n s , and could then be used to i d e n t i f y the TR vowels. Whether i t i s used or not cannot be determined i n t h i s design, and no p u b l i s h e d work on i t s u t i l i t y as a cue could be 57 found. In summary, the TR s t i m u l i have the a d d i t i o n a l cues of du r a t i o n , g l i d e , and i n t e n s i t y . D uration i n f o r m a t i o n does not appear t o have a f f e c t e d the e r r o r p a t t e r n s . Forrnant movement i n f o r m a t i o n contained i n the g l i d e s present may have aided d i f f e r e n t i a t i o n of g l i d e d and non-glided vowels. I n t e n s i t y cues are present, but whether or not they are used i s not d i s e e r n a b l e i n t h i s experimental design. Although p o s s i b l y u s e f u l , the presence of these cues (du r a t i o n , g l i d e and i n t e n s i t y ) , does not provide an adequate e x p l a n a t i o n of performance i n the TR c o n d i t i o n , e s p e c i a l l y f o r the c o n t e x t s showing performance equal t o that i n the other c o n d i t i o n s . Therefore, the remaining cue, t r a n s i t i o n s , a l r e a d y shown t o play a r o l e i n vowel i d e n t i f i c a t i o n (Lindblorn & Studdert—Kennedy, 1967; Strange et a l . , 1976; Strange et a l . , 1983; J e n k i n s et a l . , 1983), may reasonably be invoked as an e x p l a n a t i o n f o r the good performance i n the TR c o n d i t i o n . T r a n s i t i o n s , then, could be s a i d to be the major cues a v a i l a b l e i n the TR s t i m u l i . They provide s u f f i c i e n t i n f o r m a t i o n t o cue vowel i d e n t i t y and are, most of the time, s u f f i c i e n t l y robust t o do so even i n the presence of the tr a n s c o n s o n a n t a l c o a r t i c u l a t i o n a r i s i n g i n a V i C V a C V i utterance. The u t i l i t y of t r a n s i t i o n s , however, i s not uniform over a l l vowel contexts. In the presence of c o a r t i c u l a t i o n from surrounding /a/'s, the i n d e n t i f i c a t i o n of vowels cued by t r a n s i t i o n s alone i s c o n s i d e r a b l y impaired, whereas f o r vowels embedded i n / i / or /u/ contexts, the t r a n s i t i o n s succeed e q u a l l y well i n cueing the i d e n t i t y of the vowel as the i n f o r m a t i o n a v a i l a b l e i n the unmodified vowel (V a ) e x c i s e d from i t s Vt CV a C V i utterance. 6.2 Prob1em Vowels Performance on two of the t h r e e c o n t e x t s i n the CO c o n d i t i o n was high ( l e s s than 2°/. e r r o r s ) . T h i s can be a t t r i b u t e d , i n part, to experimental d e s i g n f a c t o r s such as the mon i t or i ng of prod uct i on q ua1i t y, d i a l e c t a l matching, good l i s t e n i n g c o n d i t i o n s , and the use of s u b j e c t s with t r a n s c r i p t i o n experience, a l l f a c t o r s c i t e d as important i n p r e v i o u s s t u d i e s (e.g. Macchi, 1980; flssman et a l . , 1982). Reasons f o r the decreased performance on the /a/ context i n the c o n t r o l c o n d i t i o n are not obvious, but p a r a l l e l decreases i n performance f o r t h i s context i n the other two c o n d i t i o n s . P o s s i b l e e x p l a n a t i o n s f o r t h i s e f f e c t w i l l be d e a l t with l a t e r . D i f f i c u l t y occurred with s e v e r a l of the V a vowels (those to be i d e n t i f i e d ) i n the cont ro1 cond i t ion, d e s p i t e the precaut ions taken. P o s s i b l e reasons f o r t h i s s h a l l be d i s c u s s e d i n t h i s sect ion. The monitoring of p r o d u c t i o n through speaker and experimenter judgements, as well as through unanimous agreement from t h r e e t r a n s c r i b e r s was s t i l l not s u f f i c i e n t t o i n s u r e the intended responses f o r a l l vowel types i n the CO c o n d i t i o n from a l l the s u b j e c t s . C o n s i s t e n t l y d i f f e r i n g p e r c e p t i o n s from those of the e v a l u a t o r s were obtained f o r the /a/ and /"'/ s t i m u l i i n a l l contexts, and f o r the /UY s t i m u l u s i n the /u/ context. P o s s i b l e e x p l a n a t i o n s f o r the f a c t that these items were not judged as anomalous i n i t i a l l y i n c l u d e low r e l i a b i l i t y of the t r a n s c r i p t i o n s and changes i n p e r c e p t i o n from the case where the vowel ( V B ) i s i n the V i C V a C V i context to the case where i t i s e x c i s e d from i t s context. The p o s s i b i l i t y of low r e l i a b i l i t y f o r the t r a n s c r i p t i o n s was examined f i r s t . F-our add i t iona 1 t r a n s e r i pt ions of the e n t i r e V i C V a C V i u t t e r a n c e s were made, t h i s time by four l i s t e n e r s who had p a r t i c i p a t e d i n the f o r c e d c h o i c e ~~CVa C ~ experiment. The s u b j e c t s were chosen on the b a s i s of a v a i l a b i l i t y , not on t h e i r p a r t i c u l a r performances i n the e a r l i e r experiment. Good agreement was obtained f o r the V i C V 8 C V i t r a n s c r i p t i o n s and the —CVaC— f o r c e d c h o i c e r e s u l t s i n that, high e r r o r r a t e s occurred i n the t r a n s c r i p t i o n s f o r those vowels designated p r o b l e m a t i c i n the f o r c e d c h o i c e procedure, and low e r r o r r a t e s occurred f o r the others. Examination of the i n d i v i d u a l l i s t e n e r s ' responses f o r the problematic vowels (/a/ and /'"/ i n a l l contexts, /U/ i n context /u/) showed, however, l i t t l e c o n s i s t e n c y i n t h e i r p e r c e p t i o n over the two s i t u a t i o n s (see Table XI). 60 Table XI Co rnparison of I d e n t i f i c a t i o n Responses f o r V_i PVe CV i T r a n s c r i p t i o n s and ~-CVe C- Forced Choices f o r the Prob 1 em Vowe 3, s subj. u t t . /ababa//ab Aba/ /ubabu//ub''bu//ubUbu/ / i b a b i / / i b"'bi / #1 - C V a C ~ 4/•-•/ 4/•-•/ 4/••'•/ 4 /-V 4 /•-•/ 3/a/ 4/'V V , C V a C V , /a/ /a/ /a/ /a/ /U/ /a/ /a/ #£' - C V a C - 5/-V 3/-V 5/-V 5/-V S A V 5/a/ 5/a/ V , C V a C V , /a/ /a/ /a/ /a/ /U/ /a/ /a/ #3 - C V a C ~ 4/-'V 3/a/ 3/a/ 5/""/ 4/-/ 4/a/ 4/"V V , C V a C V , /a/ /•-•/ /a/ /'V /U/ /a/ /-V #4 --CV aC- 4/-/ 4/a/ 5/a/ 4/-"V 4/-"V 4/a/ 3/a/ V . C V a C V t /••/ /ae/ /a/ /•'•/ /U/ /a/ /U/ The u n d e r l i n e d vowel i n the f i r s t l i n e i s the vowel t o be i d e n t i f i e d . The fo r c e d c h o i c e — C V a C — s i t u a t i o n i n v o l v e d f i v e r e p e t i t i o n s of each vowel type i n t h i s p o s i t i o n . O^'ly the modal response i s recorded here. In almost a l l cases, the remainder was e i t h e r /a/ i f the modal was /'"'/, or /•"/ i f i t were /a/. The V | C V a C V i s t i m u l i were only presented once each and the t r a n s c r i p t i o n s f o r each are recorded here. The s u b j e c t s i n v o l v e d i n t h i s comparison a l l had t r a i n i n g i n t r a n s c r i p t i o n . Although the /a/ and /''/ vowel p a i r i s not c o n t r a s t i v e i n the d i a l e c t of the r e g i o n the s u b j e c t s were drawn from, i t i s expected that, because of t h e i r t r a i n i n g , they should have been a b l e to- r e l i a b l y t r a n s c r i b e the phones. The presence of disagreements f o r the /a/ and /'"/ t r a n s c r i p t i o n s a c r o s s s u b j e c t s , and the i n c o n s i s t e n t nature o f the c h o i c e s f o r the f i v e r e p e t i t i o n s of each of the — C V e C — ' s by each s u b j e c t suggest that these vowel types were not produced c o n t r a s t i v e l y by the speaker, but were produced s u f f i c i e n t l y d i f f e r e n t l y t o r e s u l t i n u n c e r t a i n t y and the use of more than one symbol by some? of the l i s t e n e r s . The tendency t o attempt to f i t the vowel p r o d u c t i o n s i n more than one category was probably exacerbated by the b i a s e x i s t i n g from the l i s t e n e r s 1 knowledge that both /a/ and /""/ were a v a i l a b l e as p o s s i b l e responses. ft number of the twenty s u b j e c t s i n the — C V a C— experiment spontaneously commented that they had d i f f i c u l t y d i f f e r e n t i a t i n g these two vowel types, and that the /a/ seemed to lack "openness". For the /U/ i n the /u/ context, the s u b j e c t s had p r e v i o u s l y responded with an /a/ or an /"'/ to p r e s e n t a t i o n s of the vowel i n the - C V a C - form. In the a d d i t i o n a l V t C V e C V i t r a n s c r i p t i o n task, t h i s vowel was t r a n s c r i b e d c o r r e c t l y by a l l fo u r l i s t e n e r s . These fo u r t r a n s c r i p t i o n s , p l u s those of the o r i g i n a l e v a l u a t o r s , provide r e l i a b l e evidence that t h i s vowel a c t u a l l y had the q u a l i t y /U/ i n the f u l l V t C V e C V t form. I f so, t h i s makes the su g g e s t i o n of a. change i n vowel q u a l i t y f o r V a upon removal from the surrounding vowel ( V t ) context more t enab 1 e. I f vowel q u a l i t y changes when the vowel i s removed from i t s surrounding utterance, that would suggest that necessary vowel i d e n t i f i c a t i o n i n f o r m a t i o n i s spread over a segment l a r g e r than a C VC s y l l a b l e . T h i s p o s s i b i l i t y i s c o n s i s t e n t with f i n d i n g s on the p e r c e p t i o n of other types of c o a r t i c u l a t i o n (e.g. flli et a l . , 1371; Benguerel & Ode1man, 1977) and the well e s t a b l i s h e d d i f f i c u l t y encountered i n i d e n t i f y i n g vowels e x c i s e d from connected d i s c o u r s e as compared to i d e n t i f i c a t i o n of those produced i n i s o l a t i o n . For nine of the ten vowels i n the present study, i d e n t i f i c a t i o n of the medial vowel was not a f f e c t e d by the removal of c o a r t i c u l a t o r y i n f o r m a t i o n a v a i l a b l e i n the surrounding vowel context of a V t C V a C V i utterance. For the vowel /U/, however, the tr a n s c o n s o n a n t a l c o a r t i c u l a t i o n i n f o r m a t i o n provided by the frame /uCv"aCu/ i s a p p a r e n t l y c r i t i c a l t o i t s p e r c e p t i o n i n the V a p o s i t i o n . 6.3 The E f f e c t of Outpoint Placement The h y p o t h e s i s t o be t e s t e d i n t h i s study was that the in f o r m a t i o n contained i n t r a n s i t i o n s would not be as e f f e c t i v e as that contained i n the s t e a d y - s t a t e i n cueing vowel i d e n t i t y . T r a n s i t i o n only, s t e a d y - s t a t e only and unmodified vowel s t i m u l i were used t o t e s t t h i s hypothesis. R c r u c i a l f a c t o r i n drawing c o n c l u s i o n s from the i d e n t i f i c a t i o n of these s t i m u l i i s whether or not s t e a d y - s t a t e i n f o r m a t i o n was present w i t h i n the " t r a n s i t i o n o n l y " vowels. In chapter four, i t was s t a t e d that, f o r /U/,/u/,/oU/, and /•"'•/, d i f f i c u l t y was encountered i n determining the boundary between s t e a d y - s t a t e and t r a n s i t i o n because of the n e a r l y h o r i z o n t a l nature of the t r a n s i t i o n s . For those four vowels, the cut point placement r e s u l t e d i n s t e a d y - s t a t e i n f o r m a t i o n being present i n the t r a n s i t i o n segments of the TR s t i m u l i . For the g l i d e d vowel /oU/, the low percentage used ( 1 0 ° / ) f o r the n e a t — f l a t t r a n s i t i o n reduced t h i s concern, but some steady-s t a t e i n f o r m a t i o n was s t i l l present. Presence of t h i s " t a r g e t " i n f o r m a t i o n i n the t r a n s i t i o n segments could be expected to lower e r r o r r a t e s i n vowel i d e n t i f i c a t i o n performance i n the TR cond i t ion. Examination of the performance of /u/, /UV, /""/, and /oU/ i n the TR c o n d i t i o n i n Table X, r e v e a l s that these vowels do, i n f a c t , show lower e r r o r r a t e s than do the other vowel s t i m u l i . The vowel /'"/ d i d not show good performance, and had to be c o l l a p s e d with /a/, but t h e i r combined e r r o r r a t e was low (£"/ e r r o r s over a l l c o n t e x t s ) . The vowels /U/ ( i n co n t e x t s / i / and /u/), /u/, and /oU/ are among the best i d e n t i f i e d vowels, although the presence of the a d d i t i o n a l g l i d e cue f o r /oU/ must be i n c l u d e d i n the e x p l a n a t i o n o f good performance. Good performance on the vowels d e s c r i b e d , however, may be e x p l a i n e d by reasons other than the presence of " t a r g e t " i n f o r m a t i o n , such as the g l i d e cue mentioned, the s a l i e n t p o s i t i o n of /u/ i n the vowel t r i a n g l e (as an extreme v a l u e ) , and the lack of c l o s e neighbours f o r /•"/ because of the c o l l a p s i n g of /a/ and /''/. I f the presence of s t e a d y - s t a t e i n f o r m a t i o n i s the c r i t i c a l f a c t o r i n a i d i n g i d e n t i f i c a t i o n i n the TR vowel s t i m u l i , then any vowels d i s p l a y i n g s t e a d y - s t a t e p a t t e r n s w i t h i n the designated t r a n s i t i o n segments should be 64 expected to show c o r r e s p o n d i n g l y lower e r r o r r a t e s , while those not possessing t h i s " t a r g e t " i n f o r m a t i o n should show higher r a t e s o f e r r o r . The vowel5 showing the highest e r r o r r a t e s i n the TR c o n d i t i o n , and c o n t r i b u t i n g most to the poor performance were / i / , / I / , / E / , and /ae/ with e r r o r r a t e s of 19°/ or above. fill showed a c l e a r i n c r e a s e i n e r r o r s from the CO and SS c o n d i t i o n s t o the TR c o n d i t i o n . I f these p a r t i c u l a r vowels are examined a c r o s s contexts, i t can be seen that i n the / i / context, performance i s comparable to the other vowels and t o the other c o n d i t i o n s . In the /u/ context, performance i s poor f o r / i / , / I / , and /E/ at e r r o r r a t e s of at l e a s t 2.3"/.. Almost a l l the vowels ( / i / , /I/,/E/,/ae/,/ei/,/u/) i n the /a/ context are p o o r l y i d e n t i f i e d ( e r r o r s at or above £8°/). If one examines the degree of match between the a c t u a l r e l a t i v e d u r a t i o n s of the t r a n s i t i o n s i n Table II and the r e l a t i v e d u r a t i o n s chosen to base the e u t p o i n t s used on (Table I I I ) , the /u/ context shows the g r e a t e s t number of mismatches (items u n d e r l i n e d i n Table I I ) , with f i v e vowels (/a/, /E/, /U/,/""/, /oU/) having some s t e a d y - s t a t e i n t h e i r t r a n s i t i o n segments. These p a r t i c u l a r vowels are not, however, the vowels with the lowest i d e n t i f i c a t i o n e r r o r r a t e s . Pis w e l l , d e s p i t e c o n t a i n i n g a l a r g e number of vowels with mismatches between a c t u a l and chosen r e l a t i v e d u r a t i o n s (where s t e a d y - s t a t e i n f o r m a t i o n i s present i n the t r a n s i t i o n segments), /u/ i s not the best p e r c e i v e d context. Furthermore, i n the /a/ and / i / c ontexts, the p a r t i c u l a r mismatches o c c u r r i n g do not match up with the p a t t e r n of performance i n the i d e n t i f i c a t i o n task. Pis w e l l , the mismatches i n the /a/ and / i / c o n t e x t s are s i m i l a r i n number and pattern, yet performance f o r each d i f f e r s g r e a t l y . Pin a d d i t i o n a l source of support f o r t r a n s i t i o n s as cues t o vowel i d e n t i t y was the i d e n t i f i c a t i o n performance f o r those vowels f o r which the mismatch went i n the o p p o s i t e d i r e c t i o n , r e s u l t i n g i n incomplete t r a n s i t i o n s f o r e i t h e r the i n i t i a l or f i n a l segments. Pl comparison of the a c t u a l r e l a t i v e d u r a t i o n s of Table II and the r e l a t i v e d u r a t i o n s chosen (Table III) show almost a l l the vowels having one or both of the t r a n s i t i o n segments cut too short. Performance i n the c o n t e x t s / i / and /u/, f o r the TR c o n d i t i o n , shows no decrease i n i d e n t i f i c a t i o n r a t e s as a r e s u l t of t h i s e a r l y t r u n c a t i o n ; performance i s as good as i n the SS and CO c o n d i t i o n s . Another point of comparison i s the i d e n t i f i c a t i o n performance f o r the? two v e r s i o n s of / i / and /u/. Two d i f f e r e n t s e t s of c u t p o i n t s were used f o r the vowels / i / and /u/, based on how the vowels were grouped. The in t e r m e d i a t e grouping r e s u l t e d i n a mismatch f o r the vowel / i / i n the / i / context, with s t e a d y — s t a t e i n f o r m a t i o n present i n the t r a n s i t i o n segment, but the e r r o r r a t e s f o r the two v e r s i o n s were s i m i l a r (10"/ and 14"/). As w e l l , d i f f e r i n g degrees of t r a n s i t i o n present i n the TR c o n d i t i o n f o r the two v e r s i o n s of / i / and /u/ i n the other c o n d i t i o n - c o n t e x t combinations, as a r e s u l t of the £0°/. and £5°/» f i n a l r e l a t i v e d u r a t i o n s used, do not r e s u l t i n d i f f e r e n t performance. The presences of completed t r a n s i t i o n s t h e r e f o r e , does not appear to be c r u c i a l f o r i d e n t i f i c a t i o n . The lack of r e l a t i o n between the l o c a t i o n used f o r the e u t p o i n t s and i d e n t i f i c a t i o n performance suggest that the presence or absence of s t e a d y — s t a t e i n f o r m a t i o n does not g r e a t l y a f f e c t vowel p e r c e p t i o n , and that completed t r a n s i t i o n s are not r e q u i r e d f o r i d e n t i f i c a t i o n . 6.4 Poor Performance i n the /a/ Context Performance i n the /a/ context i s c o n s i s t e n t l y below performance i n the other contexts. The d i f f e r e n c e i s not s t a t i s t i c a l l y s i g n i f i c a n t f o r the CO or the SS c o n d i t i o n s , but i s s t i l l present. For the TR c o n d i t i o n , a s i g n i f i c a n t i n c r e a s e i n e r r o r s from those i n the CO and SS c o n d i t i o n s occurs. Poss i b1e f a c t o r s t o be cons i d ered i n c l u d e poor prod uct i on q u a l i t y , c o a r t i e u l a t o r y c o n s t r a i n t s , cut point placement and s u b j e c t - s p e c i f i c performance. The p r o d u c t i o n of /a/ i n the v"s p o s i t i o n was shown e a r l i e r to have been non—contrast i v e with /""/. It i s p o s s i b l e that the /a/ was p o o r l y produced i n the outer vowel p o s i t i o n too. Only one of the o r i g i n a l t h r e e e v a l u a t o r s took e x c e p t i o n to the c h o i c e of /a/, /u/, or / i / as a t r a n s c r i p t i o n of the outer vowel (she t r a n s c r i b e d the outer /a/ as / " " / ) , but an /a/ b i a s was introduced when a set of the intended productions was provided to the t h r e e e v a l u a t o r s . I f the /a/ vowel was 67 a c t u a l l y produced as an /""V, i t s t i l l does not e x p l a i n the high e r r o r r a t e of the "/a/" context s i n c e the t r a n s c o n s o n a n t a l c o a r t i c u1at i on wo u1d occ ur reg ard1ess of whether t h e or i g i na1 context was that of an /a/ or an /'"/. Ohrnan (1366) d i d not s p e c i f i c a l l y examine /'"/ coart i c u l a t ion, but t h e r e i s no reason t o t h i n k that i t would not occur. If, f o r some reason, i t should not occur, the e f f e c t would be b e t t e r performance i n the T R cond i t i on - / a / cont e x t (act ua 11 y / '"• / cont e x t) s i nee potent ia11y 1ess " d i s t o r t i o n " wouId be introduced. Q u a l i t y of p r o d u c t i o n should not, t h e r e f o r e , be a f a c t o r i n the poor i d e n t i f i c a t i o n performance f o r vowels produced i n the /a/ context. P l a c e of p r o d u c t i o n a f f e c t s the s p e c t r a l p a t t e r n of the vowels. There i s the p o s s i b i l i t y that /a/, as the only low vowel t e s t e d i n the outer p o s i t i o n of the V t C V a C V i utterance, has an e f f e c t on the t r a n s c o n s o n a n t a l vowel ( V e ) and i t s t r a n s i t i o n that i s somehow more damaging than the e f f e c t of the high vowels. T h i s assumes, of course, that the /a/ was a c t u a l l y produced as a low back vowel, not as the more c e n t r a l vowel /•""/. I f the /a/ was a c t u a l l y a r t i c u l a t e d , g r e a t e r c o a r t i c u l a t o r y e f f e c t s might be expected i f the c o a r t i c u l a t i o n was extreme, f o r example, from a low to a high and back to a low vowel, l i k e /abiba/. However, the e r r o r p a t t e r n i n t h i s context does not r e v e a l d i f f i c u l t y only with high vowels (see Table IX); i n the TR c o n d i t i o n , the highest e r r o r r a t e s occur f o r the high f r o n t vowels / i / and / I / , but high e r r o r r a t e s 6 8 a l s o occur throughout the vowel set- In the other c o n d i t i o n s , the e r r o r s are l e s s widespread; / i / and /!/ are s t i l l among the highest, taut /ae/ (low f r o n t ) i n the CO and / e i / (mid f r o n t ) i n the SS c o n d i t i o n s a l s o show e r r o r r a t e s as high or higher. A r t i c u l a t o r y movement from low to high to low does not, t here f o r e , sh ow a cons i s t ent e o r r e 1 a t i on with poor i d e n t i f i c a t i o n . C o a r t i c u l a t o r y d i f f i c u l t i e s stemming from the r e v e r s e p a t t e r n (high-low-high) do not r e s u l t i n h i g h e r e r r o r r a t e s e i t h e r , as evidenced by the good i d e n t i f i c a t i o n performance obtained f o r low vowels produced i n / i / and /u/ contexts. P r o d u c t i o n c o n s t r a i n t s do not appear to r e s u l t i n p e r c e p t u a l ambiguity, and do not account f o r the high e r r o r r a t e s f o r the vowels i n the /a/ context. A r t i c u l a t o r y f a c t o r s do not provide c l e a r evidence to account f o r the performance obtained i n the /a/ context. It i s p o s s i b l e that a r t i c u l a t o r y p a t t e r n s may r e s u l t i n s u b t l e a c o u s t i c changes that r e s u l t i n s p e c i f i c i d e n t i f i c a t i o n d i f f i c u l t i e s . The a c o u s t i c p a t t e r n s of the medial vowels were, t h e r e f o r e , examined f o r unusual c h a r a c t e r i s t i c s . The spectrograms f o r each vowel type v a r i e d s l i g h t l y a c r o s s contexts, presumably as a r e s u l t of the t r a n s c o n s o n a n t a l c o a r t i c u l a t i o n from the outer vowels. Measures of t r a n s i t i o n onset and o f f s e t f r e q u e n c i e s , and t r a n s i t i o n s l o p e angles were taken. There were no l a r g e or c o n s i s t e n t d i f f e r e n c e s f o r any of the vowel types, taken i n d i v i d u a l l y or as a whole, a c r o s s c o n t e x t s . A c o u s t i c a l l y , then, the context /a/ does not appear 6 9 t o s t a n d o u t f r o m t h e o t h e r t w o c o n t e x t s . P e r f o r m a n c e i n t h e / a / c o n t e x t w a s m o s t n o t a b l y i m p a i r e d i n t h e T R c o n d i t i o n . T h e s e r e s u l t s m i g h t b e a n a r t i f a c t o f c u t p o i n t p l a c e m e n t c a u s e d b y i n f l a t e d r a t e s f o r t h e / i / a n d / u / c o n t e x t s d u e t o t h e p r e s e n c e o f s t e a d y - s t a t e ? i n f o r m a t i o n i n t h e t r a n s i t i o n s e g m e n t s . T h i s p o s s i b i l i t y w a s d e a l t w i t h e a r l i e r ( s e c t i o n 6 . 3 ) i n a n a t t e m p t t o e x p l a i n t h e p a t t e r n o f p e r f o r m a n c e o v e r v o w e l t y p e s . C o n t e x t s / a / a n d / i / h a d v e r y s i m i l a r p a t t e r n s o f m i s m a t c h b e t w e e n a c t u a l r e l a t i v e d u r a t i o n s a n d c h o s e n r e l a t i v e d u r a t i o n s , y e t p e r f o r m a n c e f o r t h e s e t w o d i f f e r e d g r e a t l y . A s w e l l , t h e a c t u a l e r r o r p a t t e r n s o v e r t h e v o w e l t y p e s d i d n o t m a t c h u p w i t h t h e p a t t e r n o f m i s m a t c h e s o v e r t h e v o w e l t y p e s . T h e p r e s e n c e o r a b s e n c e o f s t e a d y — s t a t e i n f o r m a t i o n t h u s d o e s n o t e x p l a i n t h e p e r f o r m a n c e v a r i a t i o n a c r o s s c o n t e x t s . T h e d e t r i m e n t a l e f f e c t o f t h e / a / c o n t e x t o n m e d i a l v o w e l i d e n t i f i c a t i o n i s s t r o n g , b u t t h e p o s s i b i l i t y t h a t t h e p a t t e r n o f e r r o r s may h a v e b e e n c a u s e d b y a f e w s u b j e c t s a n d b e , i n f a c t , i d i o s y n c r a t i c , w a s e x a m i n e d . S u b j e c t s w e r e r a n k o r d e r e d b y t h e a v e r a g e p e r f o r m a n c e a c r o s s c o n d i t i o n s a n d s e p a r a t e A N D V A S o n t h e t o p a n d b o t t o m t e n s u b j e c t s w e r e p e r f o r m e d . T h e r e s u l t s f r o m t h i s a n a l y s i s w e r e t h e s a m e a s t h a t f o r a l l t h e s u b j e c t s t o g e t h e r , s h o w i n g a c o n t e x t e f f e c t f o r / a / , . T h u s , t h e l o w e r p e r f o r m a n c e o n v o w e l s i n t h e / a / c o n t e x t i s a c o n s i s t e n t e f f e c t o v e r t h e t w e n t y s u b j e c t s t e s t e d . N o n e o f t h e f a c t o r s c o n s i d e r e d a c c o u n t f o r t h e p o o r 70 p e r f o r m a n c e o f s u b j e c t s i n t h e / a / c o n t e x t , o r , more s p e c i f i c a l l y , i n t h e TR c o n d i t i o n f o r t h i s c o n t e x t . T h i s e f f e c t , a l t h o u g h r o b u s t , i s u n e x p 1 a i n a b 1 e by t h e i n f o r m a t i o n e x a m i n e d s o f a r . U n t i l e x p l a i n e d , t h e p e r f o r m a n c e i n t h e TR c o n d i t i o n - / a / c o n t e x t p r e s e n t s a c o n s t r a i n t on c o n c l u d i n g t h a t t r a n s i t i o n s a r e s u f f i c i e n t t o c u e v o w e l i d e n t i t y . T h i s c o n s t r a i n t becomes more s e v e r e w i t h t h e p o s s i b i l i t y t h a t t h e i n t e n d e d low bac k / a / was a c t u a l l y p r o d u c e d a s t h e more c e n t r a l /•"'•/" t h e l i k e l i h o o d o f o t h e r c o n t e x t s s h o w i n g s i m i l a r p a t t e r n s i n c r e a s e s a s t h e v o w e l i n q u e s t i o n becomes l e s s a r t i c u l a t o r i l y " e x t r e m e " . However, t h e f a c t t h a t p e r f o r m a n c e i n t h e / a / c o n t e x t was a l s o i m p a i r e d i n t h e o t h e r c o n d i t i o n s , a l b e i t t o a l e s s e r and n o n - s i g n i f i c a n t e x t e n t , s u g g e s t s t h a t t h e i d e n t i f i c a t i o n d i f f i c u l t y may n o t be due t o t h e i n s u f f i c i e n c y o f t r a n s i t i o n s a s c u e s but r a t h e r t o some e s s e n t i a l a m b i g u i t y a b o u t v o w e l s p r o d u c e d i n t h e / a / c o n t e x t , w h i c h i s i n c r e a s e d when c o n t i g u o u s c h u n k s a s l a r g e a s t h e s t e a d y s t a t e (50°/.--65°/. o f t h e t o t a l d u r a t i o n ) a r e removed f r o m t h e v o w e l . 6. 5 S u m rn a r y a n d C o n c 1 u s i o n s T h i s s t u d y h a s e x a m i n e d t h e e f f e c t o f t r a n s c o n s o n a n t a l c o a r t i c u 1 a t i o n on t h e u s e f u 1 n e s s o f t r a n s i t i o n s a s c u e s t o t h e p e r c e p t i o n o f t h e m e d i a l v o w e l <VS ) i n V 4 C V S C V t u t t e r a n c e s , w h ere V t was one o f t h r e e v o w e l s (/a/,/u/, o r / i / ) , C was t h e p l o s i v e /b/, and V a was one o f t e n v o w e l s ( / i / , / I / , / e l / , /E/, /a e / , / a / , /"V, /oU, /U/, /u/>. 71 For two of the t h r e e Vt contexts, the vowels i n -CVa C~ s t i m u l i with the V a s t e a d y - s t a t e removed (TR c o n d i t i o n ) were i d e n t i f i e d e q u a l l y as well as vowels i n -CV eC- s t i m u l i with the t r a n s i t i o n s removed (SS c o n d i t i o n ) and those i n unmodified-CV a C— s t i m u l i (CO c o n d i t i o n ) . For these contexts, i n f o r m a t i o n from t r a n . i t i ons a1one i s s u f f i c i e n t t o c ue vowe1 i d ent i t y at a l e v e l equal to that of the vowel cue formly c o n s i d e r e d to be the most important, s t e a d y - s t a t e i n f o r m a t i o n . ft major l i m i t a t i o n on c o n c l u d i n g that t r a n s i t i o n s cue vowel i d e n t i t y e q u a l l y well as s t e a d y - s t a t e or combined i n f o r m a t i o n r e s u l t s from the uneven performance o c c u r r i n g a c r o s s contexts. Vowels i n an /a/ context were i d e n t i f i e d c o n s i s t e n t l y more po o r l y than those i n /u/ or / i / c o n t exts. T h i s e f f e c t was not s t a t i s t i c a l l y s i g n i f i c a n t f o r the CO or SS c o n d i t i o n , taut was h i g h l y s i g n i f i c a n t f o r the TR c o n d i t i o n . V a r i o u s e x p l a n a t i o n s of the e f f e c t , i n c l u d i n g q u a l i t y of production, cut point placement, and a r t i c u l a t o r y c o n s t r a i n t s , were considered. None were found to be adequate. ft major d i f f i c u l t y i n accounting f o r t h i s e f f e c t was the small number of c o n t e x t s used. F u r t h e r understanding of t h i s p a t t e r n of performance r e q u i r e s a more systematic, comprehensive examination of the e f f e c t of c o a r t i c u l a t i o n by other vowel contexts. It can be concluded that, i n V 4CV aCV t utterances, where Vt i s /u/ or / i / , the medial vowel ( V a ) , with s t e a d y - s t a t e i n format i on removed, can be i d ent i f i ed we11. In t h i s s i t u a t i o n , t r a n s i t i o n s can, depending on the nature of the t ranscorisonant a 3. coart i c u 1 at i on present, cue i dent i t y equa 11 y well as s t e a d y - s t a t e i n f o r m a t i o n alone or even i n f o r m a t i o n from s t e a d y - s t a t e and t r a n s i t i o n s combined. The u s e f u l n e s s of t r a n s i t i o n s as cues t o vowel i d e n t i t y i n the presence of tr a n s c o n s o n a n t a l c o a r t i c u l a t i o n , however, cannot be g e n e r a l i z e d to a l l vowel c o n t e x t s without f u r t h e r examination of the r e l a t i v e uniqueness of the e f f e c t of the /a/ context. The reasons f o r t h i s e f f e c t are yet to be e x p l a i n e d and the range of e f f e c t s of vowel context are yet to be charted, so d e f i n i t i v e c l a i m s about the s u f f i c i e n c y of t r a n s i t i o n s alone as cues t o vowel i d e n t i t y cannot be made. Whether or not t r a n s i t i o n s are s u f f i c i e n t i n themselves to cue vowel i d e n t i t y , i t can s t i l l be concluded, with some s u r i t y , that the performance on t r a n s i t i o n only s t i m u l i s t r o n g l y supports the p o t e n t i a l u s e f u l n e s s of i n f o r m a t i o n from t r a n s i t i o n s i n the normal i d e n t i f i c a t i o n s i t u a t i o n , where s t e a d y - s t a t e and t r a n s i t i o n i n f o r m a t i o n are both a v a i l a b l e . 73 REFERENCES CITED A l i , L., Gallagher, T., G o l s t e i n , J . , and D a n i l o f f , R. (1971). P e r c e p t i o n of c o a r t i c u l a t e d n a s a l i t y . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 49, 538-540. Assrnan, P. F. , Nearey, T. M. , and Hogan, J. T. (198£)» Vowel i d e n t i f i c a t i o n s o r t h o g r a p h i c , p e r c e p t u a l and acoustic-aspects. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 71_, 975-989. Benguerel, A-P., and Cowan, H.A. (1974). C o a r t i c u l a t i o n of upper l i p p r o t r u s i o n i n French. Phonetica, 30, 41-55. Benguerel, A-P., and Adelman, S. (1977). C o a r t i c u l a t i o n of l i p r o u n d i n g and i t s percept ion. In S t r u c t u r e and Process  i n Speech P e r c e p t i o n , A. Cohen and S.G. Nooteboom (eds.), NY: S p r i n g e r V e r l a g . Bond, Z. S. (1975). Id ent i f i cat i on of voweIs e xcerpt ed from context. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 6£ (Suppl. No. 1), S101. Carney, P.J. and M o l l , K.L. (1971). A C i n e f 1 u o r o g r a p h i c i n v e s t i gat i on of f r i c a t i ve consonant-vowe1 coart i cu1at ion. Phonetica, £3(4), 133--E00. Cla r k , M. and Sharf, D.J. (1373). C o a r t i c u 1 a t i o n e f f e c t s of post—consonantal vowels on the short—term r e c a l l of pre— consonantal vowels. Language and Speech, 1.6, 67-76. D i e h l , R.L., Buchwald McCusker, S., and Chapman, L.S. (1381). P e r c e i v i n g vowels i n i s o l a t i o n and i n consonantal context. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 63(1), £33-£48. Fairbanks, G. and Grubb, P. (1361). A psychophysical i n v e s t i g a t i o n of vowel forrnants. J o u r n a l of S3peech and Hearing Research, 4, £03—£13. Fromkin, V. (1966). Neuro-museu1ar s p e c i f i c a t i o n s of l i n g u i s t i c u n i t s . Language and Speech, 9, 170-199. Fu.j irnura, Q. and O c h i a i , K. (1963). Vowel i d e n t i f i c a t i o n and ph onet i c eont e x t s. Jo urna1 o f the Aco ust i ca1 Soc i et y of Amer i can, 35, 1889(A). 74 Gay, T. (1978), E f f e c t of speaking r a t e on vowel forrnant movements. Jo u r n a l of the A c o u s t i c a l S o c i e t y of America, 6*3 ( 1 ) , dc!c!c!9. Gerstman, L.J. (1968). C l a s s i f i c a t i o n of s e l f - n o r m a l i z e d vowels. IEEE Trans. Audio E l e c t r o a c o u s t i c s , AU-16, 78-80. G o t t f r i e d , T.L. and Strange, W. (1980). I d e n t i f i c a t i o n of c o a r t i c u l a t e d vowels. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 68, 16£'6-1635. Henke, W.L. (1966). Dynamic a r t i e u l a t o r y model of speech p r o d u c t i o n using computer s i m u l a t i o n . D o c t o r a l Thesis, M. I. T. House, A. S. and Fairbanks, G. (1953). The i n f l u e n c e of consonant environment upon the secondary a c o u s t i c a l c h a r a c t e r i s t i c s of vowels. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, £5, 105-113. Jenkins, J. J. , Strange, W. , and Edrnan, T. R. (1983). I d e n t i f i c a t i o n of vowels i n v o w e l - l e s s s y l l a b l e s . P e r c e p t i o n & Psychophysies, 34 (5), 441-450. Joos, M. (1948). A c o u s t i c s Phonetics. Language, £4, 1-136. Kahn, D. (1977). Near-perfect i d e n t i f i c a t i o n of speaker-randomized vowels without consonantal t r a n s i t i o n s . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 6£, Suppl. No. 1, S101. Kozhevnikov, V.A. and C h i s t o v i c h , L.A. (1965). Speech, Art icu1at ion and Percept ion ( t r a n s l a t e d from Russian), J o i n t P u b l i c a t i o n Research S e r v i c e , U.S. Dept. Commerce No. 30 (Washington). Ladefoged, P. and Broadbent, D.E. (1957). Information conveyed by vowels. Journal of the A c o u s t i c a l S o c i e t y of America, £9(1),k 98-104. L e h i s t e , I. and Meltzer, D. (1973). Vowel and speaker i d e n t i f i c a t i o n i n n a t u r a l and s y n t h e t i c speech. Language and Speech, 16, 356-364. L e h i s t e , I., and Peterson, G.E. (1959). Vowel amplitude and phonemic s t r e s s i n American E n g l i s h . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 31_(4), 4£8-435. L e h i s t e , I. and Peterson, G.E. (1961). T r a n s i t i o n s , g l i d e s and diphthongs. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 33 (^>) , £68 d76. L e h i s t e , I., and Shockey, L. (197£). On the p e r c e p t i o n of 75 co a r t i c u l a t i on e f f e e t s i n Eng1ish VCV sy11ab1es. Working Papers i n L i n g u i s t i c s No. 12, The Ohio S t a t e U n i v e r s i t y , 78-86. Lindblorn, B. E. F. (1963). S p e c t r o g r a p h i c study of vowel r e d u c t i o n . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 35, 1773-1781. Lindblorn, B. E. F. , and St uddert-Kennedy, M. (1967). On the r o l e of forrnant t r a n s i t i o n s i n vowel r e c o g n i t i o n . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 42, 830-843. L i n t s , L.B. and Sherman, D. (1961). Phonetic elements and p e r c e p t i o n of n a s a l i t y . J o u r n a l of Speech and Hearing Research, 4, 381-396,, Macchi, M.J. (1980). I d e n t i f i c a t i o n of vowels spoken i n i s o l a t i o n versus vowels spoken i n consonantal context. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 68(6), 1636-164E. M i l l a r , J.B,,, and Ainsworth, W. A. (1972). I d e n t i f i c a t i o n of s y n t h e t i c i s o l a t e d vowels and vowels i n h-d context. Ae o u s t i c a , £7? £78—£8£. Mo 11, K. L. and D a n i l o f f , R.G. (1971) . I nvest i gat i on of t he t i m i n g of v e l a r movements during speech. Journa1 of the A c o u s t i c a l S o c i e t y of America, 50_(£) , 678-684. Neary, T. M. (1977). Phonetic f e a t u r e systems f o r vowels. PhD t h e s i s , U n i v e r s i t y of Conneticut, 1977 (reproduced by the Indiana U n i v e r s i t y L i n g u i s t i c s Club, 1978). tthman, S. E. G. (1966). C o a r t i c u l a t i o n i n VCV u t t e r a n c e s : S p e c t r o g r a p h i c measurements. Jo u r n a l of the A c o u s t i c a l S o c i e t y of America, 39, 151-168. Ohrnan, S. E. G. (1967). A Numerical model of c o a r t i c u l a t i o n . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 41, 310—320. Pet er son, G.E. (1961) . Par arnet er s o f vowe 1 q u a l i t y . Jo ur na 1 o f the A c o u s t i c a l S o c i e t y of America, 4(1), 10—29. Peterson, G.E. and Barney, H.L. (195£). C o n t r o l methods used i n a study of the vowels. J o u r n a l of the A c o u s t i c a l S o c i e t y of America, £4(£), 175-184. Peterson, G.E. and L e h i s t e , I. (I960). D u r a t i o n of s y l l a b l e n u c l e i i n E n g l i s h . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 3£ (6), 693-703. Sharf, D.J. and O s t r e i e h e r , H. (1973). E f f e c t of forward and backward c o a r t i c u l a t i on on the i d e n t i f i c a t i o n of speech sounds. Language and Speech, 1_6, 196-£06. Skinner, T.E. (1977). Speaker i n v a r i a n t c h a r a c t e r i s a t i o n s of vowels, l i q u i d s , and g l i d e s using r e l a t i v e forrnant f r e q u e n c i e s . J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 6£, Suppl. No. 1, S5. Stevens, K.N. and House, A.S. (1963). P e r t u b a t i o n of vowel a r t i c u l a t i o n s by consonantal c o n t e x t : An a c o u s t i c a l study. J o u r n a l of Speech and Hearing Research, 6, 111-128. Stevens, K. N. , House, A. S. , &• Paul, A. P. (1966). Acoust i c a d e s c r i pt i on of s y l l a b i c n ue1e i s An i n t er pret at i on i n t erms of a dynamic model of a r t i c u l a t i o n . J o u r n a l of the Acoust i ca1 Soc i et y of Amer i ca, 40(1), 1£3-132. Strange, W., Jenkins, J . J . , and Johnson, T.L. (1983). Dynamic s p e c i f i c a t i o n of c o a r t i c u l a t e d vowels. J o u r n a l of the Aco ust i ca1 Soc i et y o f Amer i c a , 74 (3), 695-705. Strange, W., Verbrugge, R.R., Shankweiler, D.P., and Edman, T.R. (1976). Consonant environment s p e c i f i e s vowel i d e n t i t y . J o u r n a l of the A c o u s t i c a l S o c i e t y ofAmerica, 60, c! 1 ~> £ cl 4. Verbrugge, R.R., Strange, W., Shankweiler, D.P., and Edman, T.R.(1976). What i n f o r m a t i o n enables a 1 i s t e n e r t o map a t a l k e r ' s vowel space? J o u r n a l of the A c o u s t i c a l S o c i e t y of America, 60(1), 198~£1£. Appendi x 77 Inst r u c t ions t o Sub,ject We're loo k i n g at how people p e r c e i v e vowels, and what p a r t s of the sound are used i n vowel i d e n t i f i c a t i o n . You are going to be presented with e l e c t r o n i c a l l y modified speech sounds. L i s t e n t o the sound c a r e f u l l y , and c i r c l e the keyword on t h i s answer sheet corresponding to the vowel you t h i n k you heard. There are no r i g h t or wrong answers i n t h i s task, j u s t c i r c l e the keyword that you think. best approximates the vowel you heard. Guessing i s f i n e , j u s t be sure to choose one answer every s i n g l e time. Do not take too long to d e c i d e because t h e r e i s only f o u r seconds between sounds. I w i l l go over the coding system with you before we s t a r t . There w i l l be t h r e e s e s s i o n s of approximately f i f t e e n minutes each. One w i l l be 155 items long, and the o t h e r s w i l l be 185 items long. There i s a pause a f t e r every tenth s t i m u l u s at which time you w i l l hear a beep. The beep should c o i n c i d e with the gap on the paper. I f i t doesn't at any time, say "stop" and I w i l l stop the tape and f i g u r e out where you have gone wrong. Before we s t a r t , I'd l i k e t o ask you some q u e s t i o n s and have you read and s i g n t h i s consent form. The only r i s k i n v o l v e d i n the experiment i s p o s s i b l e boredom. 78 •- go t o Q u e s t i o n n a i r e and Consent form (pen) -I am now going t o play ten sounds t o f a m i l i a r i z e you with the task. I am not s c o r i n g these; i t i s j u s t t o get you used t o moving a c r o s s the sheet and l o c a t i n g the answers. Here i s the answer sheet. The p r a c t i c e items are on the back page. The response s h o i c e s are w r i t t e n o r t h o g r a p h i c a l l y f o r those who do not know the phonetic alphabet. For those o f you wh do, k you w i l l probably f i n d the o r t h o g r a p h i c r e p r e s e n t a t i o n c o n f u s i n g i n i t i a l l y . That i s why I have w r i t t e n the corresp o n d i n g phonetic symbols at i n t e r v a l s on the response sheet. The vowel key i s as f o l l o w s : / b i b / , / b i b / , /beib/, /baeb/, /bab/ as i n "baaa" that a sheep would say, /b"'b/, /boUb/, /bUb/ as i n could, /bub/. Say the sounds t o y o u r s e l f , then t e l l me when you are ready. — Do p r a c t i c e items — Hny q u e s t i o n s ? O.K., now we' II s t a r t . The sounds you hear may not sound e x a c t l y l i k e those you heard on the p r a c t i c e tape. 7 9 A p p e n d i x ( c o n t . ' d )  Quest i o n n a i r e S u b j e c t Name D a t e A s s i g n e d Number O r d e r o f C o n d i t i o n s * c o d e answer s h e e t s * 1„ Have you t a k e n any c o u r s e s i n p h o n e t i c s o r have you e x p e r i e n c e i n t r a n s e r i pt i o n ? S.u I s y o u r n a t i v e t o n g u e E n g l i s h ? 3 . What o t h e r l a n g u a g e s do you s p e a k f a i r l y f l u e n t l y ? 4. What r e g i o n were you b r o u g h t up i n , o r s p e n t a m a j o r p a r t o f y o u r 1 i f e ? 5 . Do you have normal h e a r i n g ? 6- C o n s e n t form 7. Note e x c e p t i o n a l o c c u r r e n c e s 81 Appendix (cont.'d) Response Sheet' Cond. 1 £ 3 N-: / i / 1. beeb £. beeb 3. beeb 4. beeb 5. beeb 6. beeb 7. beeb 8. beeb 9. beeb 10. beeb 1 i . beeb 1£. beeb 13. beeb 14. beeb 15. beeb 16. beeb 17. beeb 18. beeb 19. beeb £0. beeb / x / b i b b i b b i b b i b b i b b i b b i b b i b b i b b i b / e i / babe babe babe babe babe babe babe babe babe babe beb beb beb beb beb beb beb beb beb beb -/ae/ bab bab bab bab bab bab bab bab bab bab /a/ baab baab baab baab baab baab baab baab baab baab / A / bub bub bub bub bub bub bub bub bub bub /ou/ bo be bo be bo be bo be bo be bobe bo be bobe bobe bobe /u/ buub b u u b buub b i.i u b b u u b b u u b buub b u u b b u u b b u u b b i b b i b b i b b i b b i b b i b b i b b i b b i b b i b babe babe babe babe babe babe babe babe babe babe beb beb beb beb beb beb beb beb beb beb bab bab bab bab bab bab bab bab bab bab baab bub baab bub baab bub baab bub baab bub baab bub baab bub baab bub baab bub baab bub bobe b bobe b bobe b bobe b bobe b bobe b bobe b bobe b bobe b bobe b /u/ ob b bo b b bo bo bo bo b bo bo ob :.b :.b ob b b b b / i / / i / / e i / /e/ /ae/ /a/ / A / / O U / / U / / U / .tb b-.ib b. .ib bi: .ib bo .ib b .ib bo .ib bo .ib bo .ib bo .ib bo ob o to ob ob ob ob ob ob ob ob *The s p a c i n g of the t e n c h o i c e s has been reduced from that presented t o the s u b j e c t s t o f i t margin demands. 

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097195/manifest

Comment

Related Items