<%BANNER%>

Effect of Syntactic Structure on Speech Production in Adults Who Stutter

Permanent Link: http://ufdc.ufl.edu/UFE0021429/00001

Material Information

Title: Effect of Syntactic Structure on Speech Production in Adults Who Stutter
Physical Description: 1 online resource (138 p.)
Language: english
Creator: Haj-Tas, Maisa Atef
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: complexity, fluency, persons, priming, stutter, stuttering, syntactic, who
Communication Sciences and Disorders -- Dissertations, Academic -- UF
Genre: Communication Sciences and Disorders thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This study examined (a) the effect of syntactic structure on speech fluency and speech reaction time (SRT) in adults who stutter, (b) the effect of syntactic priming on fluency in those participants. Fourteen persons who stutter (PWS) and fourteen persons who do not stutter (PWNS) participated in the study. A sentence structure priming paradigm was used to elicit sentences of varying syntactic forms (i.e., transitive, dative, and two-clause). The results of the fluency analyses indicated that (a) PWS produced significantly more ?repetitions and prolongations? per response than PWNS, (b) all participants produced a comparable number of ?other? types of disfluencies (e.g., interjections and revisions), (c) all participants seemed to produce more fluent responses following transitive pictures and fewer fluent responses following two-clause pictures. The results of the priming analyses indicated that (a) the presence of passive primes did not significantly increase the probability of using passives in the response, (b) the structure of the dative primes affected the probability of using those structures in the response differently between the study groups. Specifically, the PWS produced significantly more prepositional-dative responses following prepositional dative primes than following object-complement primes; however, no significant differences among the effect were observed in the PWNS. This finding was taken to suggest that priming dative sentence forms for the PWNS might be more driven by the structural aspects of the available options and for the PWS by the thematic aspects, and (c) the presence of the embedded primes did not significantly increase the probability of using embedded structures in the responses. The results of the SRT analyses indicated (a) no significant differences between the two groups in SRT, (b) that the syntactic structure of the response did not affect SRT differently between the two groups, and (c) all participants produced transitive responses at a significantly shorter SRT than dative responses, and dative responses at a significantly shorter SRT than two-clause responses. The results of the SRT were taken to suggest that increasing the syntactic complexity of the responses may influence the time both the PWS and the PWNS may need to generate such responses.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Maisa Atef Haj-Tas.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Logan, Kenneth J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021429:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021429/00001

Material Information

Title: Effect of Syntactic Structure on Speech Production in Adults Who Stutter
Physical Description: 1 online resource (138 p.)
Language: english
Creator: Haj-Tas, Maisa Atef
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: complexity, fluency, persons, priming, stutter, stuttering, syntactic, who
Communication Sciences and Disorders -- Dissertations, Academic -- UF
Genre: Communication Sciences and Disorders thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This study examined (a) the effect of syntactic structure on speech fluency and speech reaction time (SRT) in adults who stutter, (b) the effect of syntactic priming on fluency in those participants. Fourteen persons who stutter (PWS) and fourteen persons who do not stutter (PWNS) participated in the study. A sentence structure priming paradigm was used to elicit sentences of varying syntactic forms (i.e., transitive, dative, and two-clause). The results of the fluency analyses indicated that (a) PWS produced significantly more ?repetitions and prolongations? per response than PWNS, (b) all participants produced a comparable number of ?other? types of disfluencies (e.g., interjections and revisions), (c) all participants seemed to produce more fluent responses following transitive pictures and fewer fluent responses following two-clause pictures. The results of the priming analyses indicated that (a) the presence of passive primes did not significantly increase the probability of using passives in the response, (b) the structure of the dative primes affected the probability of using those structures in the response differently between the study groups. Specifically, the PWS produced significantly more prepositional-dative responses following prepositional dative primes than following object-complement primes; however, no significant differences among the effect were observed in the PWNS. This finding was taken to suggest that priming dative sentence forms for the PWNS might be more driven by the structural aspects of the available options and for the PWS by the thematic aspects, and (c) the presence of the embedded primes did not significantly increase the probability of using embedded structures in the responses. The results of the SRT analyses indicated (a) no significant differences between the two groups in SRT, (b) that the syntactic structure of the response did not affect SRT differently between the two groups, and (c) all participants produced transitive responses at a significantly shorter SRT than dative responses, and dative responses at a significantly shorter SRT than two-clause responses. The results of the SRT were taken to suggest that increasing the syntactic complexity of the responses may influence the time both the PWS and the PWNS may need to generate such responses.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Maisa Atef Haj-Tas.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Logan, Kenneth J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021429:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101206_AAAAEY INGEST_TIME 2010-12-07T00:54:26Z PACKAGE UFE0021429_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 1279 DFID F20101206_AACVYD ORIGIN DEPOSITOR PATH hajtas_m_Page_009thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
80cba455dec9b8992c3af538e14fbbe1
SHA-1
9c141b31fba78f13c0f7eb2bac0be1121278f1e8
54492 F20101206_AACUTY hajtas_m_Page_115.jpg
fab262795c58ae133160a8a660c8c6b2
b0d087bd820deb6f98f409262953d39f3256ca00
1053954 F20101206_AACVAH hajtas_m_Page_039.tif
ebeae9a0c8f32e7d0917da72a97a13bb
3a66706b82d344b932693ca17db2ce628c3ec533
6586 F20101206_AACUVB hajtas_m_Page_003.jp2
486d74321da1ccd31d0e886196b4a606
e9c36fe15a9e27f50d307ab66ef3f65963d4b713
6876 F20101206_AACVYE hajtas_m_Page_029thm.jpg
db375088b27f9be40f82e9947b8969b6
2b92a215cdac76ef3eea7ee8a263d6778fa5d917
18000 F20101206_AACVXP hajtas_m_Page_115.QC.jpg
d2fd279e4ec3f89f8b940ca484489caa
47ee788c0efeeec2be305f3209888ae3bfd3a26e
1621 F20101206_AACUTZ hajtas_m_Page_001thm.jpg
5558c94d6904cd41c51ea51945137624
42ae0e578cac3c1135e35be58b2af675c9f39a11
1051958 F20101206_AACVAI hajtas_m_Page_027.jp2
104b413c61fc487e204ff73eb93476ea
569d977aa728d02d6fef91e58fd501027e8574d9
2241 F20101206_AACUVC hajtas_m_Page_051.txt
8f305c4d981038d108d4ea571524ba00
34088563b8924a377ac7ab6b6bddda2924765855
1098 F20101206_AACUUN hajtas_m_Page_091.txt
f05c7a2fceb970cb3b0692ffff8d6145
7c6c9391a7fd144ece594c276abba3d4dfe43d9b
4630 F20101206_AACVYF hajtas_m_Page_127thm.jpg
8a3dd6904cec608cc15204ac208efac0
c18c3aae049d2b2d897fa30f946acc65d5956f17
6862 F20101206_AACVXQ hajtas_m_Page_099thm.jpg
028d427ee7012ba58e8a00babd4c10ac
bcc2ed6a3ac66d8b87d9180ebf0bcae6370f17a7
54342 F20101206_AACVAJ hajtas_m_Page_013.pro
9c181971b1aa1eebcca1270f185cce15
a0eb11b5e314715f328ed0608477677d555b570f
132959 F20101206_AACUVD hajtas_m_Page_135.jp2
900ac82d035fae23a0892afadcdd407d
c99d44017ac4878545503be2e8df8e8363171bb8
2033 F20101206_AACUUO hajtas_m_Page_076.txt
ff5c585ff97fafe32aff616e4c9d3b17
84c0a91c5bc5384c4b137c1bd1dcd384cd2c83c2
28098 F20101206_AACVYG hajtas_m_Page_020.QC.jpg
7dacb28262015ba5da6af4c7270e9970
5280dbaceef826b47d6f8e2dae9fd56f479a42c5
17437 F20101206_AACVXR hajtas_m_Page_120.QC.jpg
b7bcd37f661fb390c708dd2721a96c84
e197aac2cfa828a90dd46fd0261c15c500b84325
605487 F20101206_AACVAK hajtas_m_Page_062.jp2
4c8e6288c828930e6bf5fb271e14a8d1
26aa4690261432be14fc2bc8b0913446bb4cdfa7
25271604 F20101206_AACUVE hajtas_m_Page_006.tif
27de984ad7cf9a1c0b61767aa1ba78b4
9bdb4c5337baf71a031b550f3e3cd9f2678c2df5
106470 F20101206_AACUUP hajtas_m_Page_045.jp2
056716a6abdfd85f3b11904a775ce31f
2d3df7f1a8693ae01089268dedb58e3c0820963e
4528 F20101206_AACVYH hajtas_m_Page_039thm.jpg
99cb5c5f699ccee2ee6578bcf1f1ce64
c15807e6dd8e665878d5408561235ab56b191c57
29424 F20101206_AACVXS hajtas_m_Page_078.QC.jpg
dfbda966b60a9b250802beec9e7f5b22
de55216c9032a62adb6ef49b65019be8bf6a8318
2239 F20101206_AACVAL hajtas_m_Page_023.txt
d8ea90f585ac956d96e0beaa1c00be53
6343f691a4aad7569f73890b89930807e263d472
104333 F20101206_AACUVF hajtas_m_Page_122.jp2
c5b7238e47d17506764e2fc9b53b29d6
ae7c5111c3738209670fa980636901989c8aa11d
F20101206_AACUUQ hajtas_m_Page_035.tif
b43ef5fe10c590de7b2ebe218a1b08de
df9aaa6afd391794ba5851ae9d11b8a1442e320d
6676 F20101206_AACVYI hajtas_m_Page_044thm.jpg
582b42cc5653866915ddc468401a7b48
2c5b3d03635abd762a6eeb29964d9afdcab94104
6648 F20101206_AACVXT hajtas_m_Page_052thm.jpg
ac0ef432186b1808a7512a507f4daea8
898be2f0c11a5a252177f847405ebdb1d166b90e
90484 F20101206_AACVAM hajtas_m_Page_013.jpg
b34ad6d4f62286dc2b76a65055a8ff05
8ae4121294faf3b49544823532517d704724e643
52773 F20101206_AACUVG hajtas_m_Page_072.pro
13b9258829e04c0ab20607014f488a9c
b722f0e3b4ecbdcfda07d5c722557ef6dcaaa7f0
6590 F20101206_AACUUR hajtas_m_Page_100thm.jpg
1046825be83085af51d3990a861f123e
29ce2aa1d621aff5ec3bbf1f28ec07b17050de6a
6784 F20101206_AACVBA hajtas_m_Page_056thm.jpg
cf584ad1aca93abc3f67693567503ab5
278489f315e4ffefb470aecff59bb39784f2607d
27650 F20101206_AACVYJ hajtas_m_Page_071.QC.jpg
c773dc863cafec36ab3037652ecf1f50
102e27e9034cef520033fb060c88be9192ae170a
27741 F20101206_AACVXU hajtas_m_Page_094.QC.jpg
9b7a55a689de12bc7e640057147130d0
f699c3a147a7a0b11029e81ba6eb36c0c775db0f
120063 F20101206_AACVAN hajtas_m_Page_101.jp2
e57e776541508ed586a69196e8566984
3e0fb884ba92d08ed66d1592873763f5ef3d3dbf
430 F20101206_AACUVH hajtas_m_Page_089.txt
ab7bcc6b2638b58eb283716ccd564960
b9e3c4c0347336fa03d1f6a70d9356f04c6e600b
88223 F20101206_AACUUS hajtas_m_Page_014.jpg
d3683a602c8beb2600eb845e1135ec46
bcada0a4afeb70d524c644715f0839fba8734d96
3434 F20101206_AACVBB hajtas_m_Page_009.QC.jpg
cdcfbfa5294bffca05c68031aff205a0
35d835a471e4a969ce5be7866c778d2b8b773bc2
6789 F20101206_AACVYK hajtas_m_Page_078thm.jpg
925ab88a748d2a7bb5bb0d94737ffde7
257fe953ba34efcfe07e5e2f254945744a0d7944
26962 F20101206_AACVXV hajtas_m_Page_075.QC.jpg
f1a8b946f2bccc1ffe88f17e04479b2d
4c054ceb61f60f4e7f521239d1f7bf029832ff01
56958 F20101206_AACVAO hajtas_m_Page_132.pro
d061c310e442585311764ec5b375daa1
a7c91aae981617521672644db95a287cb48ff5fc
89304 F20101206_AACUVI hajtas_m_Page_010.jpg
67bd64df89514de80c5136a82dbf487b
da841777ae7f97256f098722662ea4c58e9072d9
1132 F20101206_AACUUT hajtas_m_Page_115.txt
32845ff1b947c83469fd77c3ef691d0c
92d63dbb20e1a0636e24956c0b85667752e34e40
14537 F20101206_AACVBC hajtas_m_Page_062.QC.jpg
f51239c398c0bd03531149d68f3c4122
1ce53f9265ae06dbd3195a73cc504c17d4ad9a1e
21064 F20101206_AACVYL hajtas_m_Page_124.QC.jpg
9767dcd423146140aec049569d110564
df344a2f49b34c56dcde1a26f107261fd1b6d910
6380 F20101206_AACVXW hajtas_m_Page_065thm.jpg
00c1b0e74767ee8d1bba11911701b776
bb52e0d85c308671fb7ffb7cf03b0a6c960a0030
93720 F20101206_AACVAP hajtas_m_Page_097.jpg
16cf8c91d1f892257dfc83fcd18bd3ab
cdf3af6264645470f10fca92afefc0fdf8d78d83
6656 F20101206_AACUVJ hajtas_m_Page_110thm.jpg
6520d035b2100f6b31bec350202ece08
7ad88c3f53073376ce4cd1b3e8fea2e97cbade26
55849 F20101206_AACUUU hajtas_m_Page_033.pro
eaa53635c2bfd03800d083085b30be4e
d4c8e5c66175276b1836c5b672d25d10c6488801
F20101206_AACVBD hajtas_m_Page_094.tif
5c623a37c35c25aec8206da76d5e5f2e
ae1487caa1ecb8fabd9c20c3a41f884ed3afcc09
29399 F20101206_AACVZA hajtas_m_Page_136.QC.jpg
803804e6132cf42f6cc64d066cc21709
5b60d10cd4a9bd9c0dfd7177d78e382bf9b17c0d
22218 F20101206_AACVYM hajtas_m_Page_123.QC.jpg
f792aa489a3e4e0daba7b08fbf099920
c91801706acab115db61ce3123fcae952a7ed3f2
28776 F20101206_AACVXX hajtas_m_Page_019.QC.jpg
a642af07b95f257a127cf0d50c58669a
9dacb29ea17f32a42c9e00cdc217a04118172827
24662 F20101206_AACVAQ hajtas_m_Page_089.jp2
15a35b259db050a493287f816fe59a7f
a0c20251f5d83b47d1a3715c7fa91d39f882d997
4049 F20101206_AACUVK hajtas_m_Page_091thm.jpg
6b7df4d6ddb59b71b7cac74f181e18ae
95f23217f2c7ce41edc8ca54fd03053836ee147d
6759 F20101206_AACUUV hajtas_m_Page_098thm.jpg
40b6b452ef6b8e5cb40a40794964df99
c629e7867115d8b9d2d5332ed57c5bf8ff5511a2
5004 F20101206_AACVBE hajtas_m_Page_113thm.jpg
65d17ae93bed97d38392b84ec79c6e21
040467b0a0a173c868d8b3e9058daa6fe1b70ddb
6919 F20101206_AACVZB hajtas_m_Page_104thm.jpg
7ffa85ef65c393fe031e75e1e2a03e48
4c97b34c3daf39ff05739277798ecfd48c455ec3
25285 F20101206_AACVYN hajtas_m_Page_045.QC.jpg
af610432e33fa0b0899518432da1f222
c847da1c9577fd7ac9b7729502da6c7b26d94e8c
4636 F20101206_AACVXY hajtas_m_Page_119thm.jpg
c109a33a08c7d27ac4457d8ac43e2548
5fe45c587593bd02aee89fc3fde3409f03fa5bbd
50860 F20101206_AACVAR hajtas_m_Page_004.pro
75e6bd942859edd28ea796038da3dec6
7f8151acf9e8f61eaa6aa209c50ee514720d8e77
F20101206_AACUVL hajtas_m_Page_121.tif
b8c6ac3ed0bb2bacfecfcca4700e2d5e
6ac032ddd4692be7a35ad6c00b44d01402253087
4735 F20101206_AACUUW hajtas_m_Page_116thm.jpg
e68fc1146a7263b0d5457c940051e720
278d0cc9dc394ba09a14a4dff01dfb06dab932d0
120595 F20101206_AACVBF hajtas_m_Page_022.jp2
2e47d1ad1022f5cda6c8c7aafa74a71b
61967b97c531608df3363db7ef9a176c5d64d4bb
4025 F20101206_AACVZC hajtas_m_Page_087thm.jpg
f7b12f88391c0d9b38cf582e5dca659e
2b35eb17b423eaa55964c03fa3c3b232647aa98d
29027 F20101206_AACVYO hajtas_m_Page_080.QC.jpg
96b2c518ec031364c4850ca21bf1a348
6ffdd7afd77d1ec2c1cffa148bbb26163d1abb1b
6736 F20101206_AACVXZ hajtas_m_Page_014thm.jpg
fe79b012d6decd4b7c9c777fe9d7bcef
eacf1ecaa365b071f1d1118f51331c404fff9022
F20101206_AACUWA hajtas_m_Page_099.tif
014e2dd97b6db29ddfc91b827f8be836
df9a55a159d434c86fdb77be39a2079dc0732e2c
F20101206_AACVAS hajtas_m_Page_092.tif
231ec90865ad9ba110758fde31d0702f
e02c407378a41c661e6d723fa1e0d16860e86199
2335 F20101206_AACUVM hajtas_m_Page_129.txt
fd97cb32b9bd0cad6b3452e92979e74b
b32c52095afd401f91480046bc6c3b1c14d76e93
F20101206_AACUUX hajtas_m_Page_076.tif
7125e63c382b1c32651d9763f65beb5f
3ce6850ee99c18b3d3d21b44a720b88971a1b90b
F20101206_AACVBG hajtas_m_Page_117.tif
9e84facd68c844fe2470afa8a39bfce7
f4845a73e7c76951030ba05ce75cd36d85d324d8
6787 F20101206_AACVZD hajtas_m_Page_135thm.jpg
40747a0d08da4e7dfbed7c37c2bb0907
015731783d0719935171c8fc2cf028c2989aa7df
26862 F20101206_AACVYP hajtas_m_Page_004.QC.jpg
22f370703d415c7aec084c663f383116
1bea8340573937c71ffb8e3a1e42387402351078
25694 F20101206_AACUWB hajtas_m_Page_069.QC.jpg
eed7d344b98e0504de0a85bde30f2805
8252a49298e36b5c4402241fd0b2860ee3dd6f2d
F20101206_AACVAT hajtas_m_Page_115.tif
e08e49ef21590b236c8c945d85d6a9f2
7383eca7e3824d3d84451170ddae44a898ce6248
54378 F20101206_AACUVN hajtas_m_Page_019.pro
58f4412613b30ae837db8dfc27041823
2abdfb5abe779607e79d033d5e74f5eb21635655
29232 F20101206_AACUUY hajtas_m_Page_037.QC.jpg
ba10180ef92afbae4c28ffc83360a80d
9387543f9498a9c7f92e4e38e663f16e4216c9d6
2234 F20101206_AACVBH hajtas_m_Page_099.txt
f2e16116f4f1d49b33a6592523601f16
cbb14e59a9c1b6d453ae51b0955a1fcde7e2f491
6890 F20101206_AACVZE hajtas_m_Page_021thm.jpg
0185d3f5835c4d5939c1bcf53cb7fea7
44ee1292f62ca991476dfdd97e1eb05772a86147
83150 F20101206_AACUWC hajtas_m_Page_057.jpg
58ddb991a13ca09f6d82afa28f94cf9b
5356379795db7acd9668342f6b8cfeb9f90c8158
13974 F20101206_AACVAU hajtas_m_Page_012.QC.jpg
7e77ad07fd1a5e135f46a0943c4a2f72
38a2e4bc106f1dfc277ef772464849a3c4b0e9b9
121755 F20101206_AACUUZ hajtas_m_Page_080.jp2
bac7dcaa16d8d85d5224aafe1791b5fd
0f7d9fccecc21152a0b6fa982fa5a98ba4530fe9
42740 F20101206_AACVBI hajtas_m_Page_121.pro
51a9287c288fb72c97210ddc9200260d
3dbbb956ca00459ea56602780c0405c608ddbb98
5501 F20101206_AACVZF hajtas_m_Page_138thm.jpg
95fa759df98e8612f92ce42b16ca1d18
eb900dc126d3158a80e8ac2a32c2a1599cea6099
3449 F20101206_AACVYQ hajtas_m_Page_012thm.jpg
4e72e13ff5d0dc843efbfa4e981da290
5c3dd42b7366384710343ebbcf4dea82aa7e9881
94659 F20101206_AACUWD hajtas_m_Page_029.jpg
40c61da17d06aaa93b5340a3449935ad
2f11a9cf73ce252b84dd1f4f67d18cc3c60d0b37
6960 F20101206_AACVAV hajtas_m_Page_103thm.jpg
551109cc5e9ac027991655f05053f9e9
5a79a44874a5f9f65e2c8a21a9e5236f7c62b4e1
6112 F20101206_AACUVO hajtas_m_Page_045thm.jpg
e7ebe81d63675c33288b71a5c2d5f91c
c1314502fd6429e8267c7930488a2a146815f8cd
F20101206_AACVBJ hajtas_m_Page_138.tif
5cfe119193f86311d79ac5b9c2c23b39
cbbafc80d8c6dc10ce6637c09d3aaceaa3f458d7
28724 F20101206_AACVZG hajtas_m_Page_051.QC.jpg
7e28db7906684c6350195fba0ba395f4
fb1893e71691b5ecf53e6a44b2272d1912b44893
6922 F20101206_AACVYR hajtas_m_Page_082thm.jpg
0d6ed920ed5659cc94d5733fa53424b4
8b8d3c33624cafe3071c69c47be7540b29f0faad
F20101206_AACUWE hajtas_m_Page_082.tif
da434812f23a39e6863756125e705abb
a656f3dc9255d9d6c71c02608f8de34449c52882
2119 F20101206_AACVAW hajtas_m_Page_094.txt
95dd9e00af76bcf1831699c65c04e635
e1f6f2fefd40aed77849776a06ceeff358e09d29
24435 F20101206_AACUVP hajtas_m_Page_001.jp2
c9c0c4a1c766948e1a81b33f7a0da628
f18ff70793271de6d39a792e3205112bf8c95aa2
50220 F20101206_AACVBK hajtas_m_Page_125.pro
3c9bcdbc84f1e19c24c6f819ac3b51ae
71f1c4e491ce76ef51136351347ef2fb17dbdb6c
23827 F20101206_AACVZH hajtas_m_Page_011.QC.jpg
b41cdf94f2aa112aa53ef61a48dd5884
a66f5941ee6b6de241459824f1b534ae1a1efb04
6263 F20101206_AACVYS hajtas_m_Page_077thm.jpg
f47616d855077df5124ed7a4430035a9
16ba48e6a52a95919b3af2f265153723eda7309f
115333 F20101206_AACUWF hajtas_m_Page_100.jp2
b11b2b28250edb502745855eb716d596
c397cd0b78d8352ed69dff2b81b0833b3f91c020
25168 F20101206_AACVAX hajtas_m_Page_067.QC.jpg
b6fd02dc9563274d0c49b861997df5d5
c62245ffb4c12c246288f8d58f5c4404595ac85d
122815 F20101206_AACUVQ hajtas_m_Page_105.jp2
06442300de4c173c2d275ff724e2a5b9
3a260f30edd6a6346ec09a61b5a43d479f9a3a4b
52420 F20101206_AACVBL hajtas_m_Page_071.pro
25bc22eac5a41c4c730d379e859980b0
140ad4799f89b1b2de0d10c5ff25eef76431db6c
29542 F20101206_AACVZI hajtas_m_Page_025.QC.jpg
4947a94aa2fcec5364bb4caffaf645c8
d9d9809530d03dfee41545098cbf9a4a603c06cf
28727 F20101206_AACVYT hajtas_m_Page_073.QC.jpg
7a9e75abcdc57b1fbe0f0fcb1bef7baf
2c8cafae613d989f1f25c8022310ac9e4ce8980e
86242 F20101206_AACUWG hajtas_m_Page_081.jpg
565ef613d073ade1d602e9035e45dcf8
506f4a4ea605eebf45584c9c2e9786b6d6589086
F20101206_AACUVR hajtas_m_Page_017.tif
511263f3a64d275b85661924ef7a68d5
b7c1101989b61e3bbb72ec7d5c4dd5ad76c76e59
23817 F20101206_AACVCA hajtas_m_Page_133.pro
13f7f066e4aa0b0d51b75dd000f15481
c427d1d17a3e7e741c58a69e485dd9d727f452be
37517 F20101206_AACVBM hajtas_m_Page_061.jpg
0421453a1998729cc3399330c8b5c37f
2e25b1e55abf3a9a5c9adaf2cb8f9904ed9d5c53
6368 F20101206_AACVZJ hajtas_m_Page_086.QC.jpg
1d9c8ea188e792f62ce01ea78a637762
467c7e127fcc6ba4daf35615ab15af038aee4766
11722 F20101206_AACVYU hajtas_m_Page_061.QC.jpg
7f295fb21ed9552d64d4b1664ce63a0a
45c5f725dbad2d23ae33749844700da59c0b8f7b
14231 F20101206_AACUWH hajtas_m_Page_063.pro
4bb8fe24b2d58ede2d81656bb630c453
d2263a5289a7e6ab6277c3f546aab28bb9a6b29b
4534 F20101206_AACVAY hajtas_m_Page_125thm.jpg
6a93adc4b4611c3157bb2e578276b030
28b50823485a4fe3a49641c9f2b3fcba5ef6f2c1
98290 F20101206_AACUVS hajtas_m_Page_040.jp2
9e389578d7dda63744c767a4e34b8ef5
4bd2a05a0f5c2294be3b16252632cc2366d01bb3
F20101206_AACVCB hajtas_m_Page_084.tif
94d1f38b5c555e6a4292d2fc2ed5eaba
3353681ffbd048e158d74341f4d33a4c9d9eba30
119693 F20101206_AACVBN hajtas_m_Page_110.jp2
f4a31d6d39570a6c6699deb670d8ff7b
2d4c05304c5dd5b1a3c84b2514eb1920d93096d2
206515 F20101206_AACVZK UFE0021429_00001.xml FULL
0c33ac26dd77913b5f466acc5fc75ff4
46dd6731700862b7f962de68d2a924986ad4e041
6330 F20101206_AACVYV hajtas_m_Page_004thm.jpg
41b0c32ed8060fc9d38772753e552947
cf0532d0da6f7b8f4399b6944bcdab47919a763a
114346 F20101206_AACUWI hajtas_m_Page_018.jp2
cef99d6af780429de9cf6595ed87cefc
5c436a762a0e4dcb52ec96fe922c3598345a0aa9
61644 F20101206_AACVAZ hajtas_m_Page_135.pro
d2b8a43be76942532bfa464a72655c84
3ba4e18ea53e93c0bee7d69a6be81090d02d00a3
123657 F20101206_AACUVT hajtas_m_Page_097.jp2
8e1062fd95535472de8e5641d2f6756a
24f6695924262ad1414e46b1d788d92a5bd32106
2028 F20101206_AACVCC hajtas_m_Page_112thm.jpg
759ba5eb9fc7e775a5c487dd97db7453
37a4f44deabb1882aef2f78a86cb3e8d6c0f6d2e
70299 F20101206_AACVBO hajtas_m_Page_121.jpg
859a9078fcf7493111407c7340beb04a
c1ae38fd17e94f10eb0eb6a232152ebda1399bc1
996 F20101206_AACVZL hajtas_m_Page_008thm.jpg
3db875ab54d340c24d5ad377fbb5d15e
2415a1df5d3faf6e473ee4aa7ef5e41cf892d5d8
6868 F20101206_AACVYW hajtas_m_Page_031thm.jpg
9d0e8fe8f545ec9839513ec720eca546
f9c9141cb439f6f462f2a44aa22c82cd141cdae0
2088 F20101206_AACUWJ hajtas_m_Page_028.txt
58197541011834db24b0129fc88aeae9
731db6dc57f1609c0cd75133b748e4cdf71a6617
95332 F20101206_AACUVU hajtas_m_Page_138.jp2
38cae1c1fefe844d7963349dc29a714c
2152826b32ca6e2e48c377de923df7cafbbd152f
28576 F20101206_AACVCD hajtas_m_Page_033.QC.jpg
cef1fc007ab0c1126c8716ebf77b0c62
cb8c6f3daef856f8486a9a9589b466708a640e80
F20101206_AACVBP hajtas_m_Page_067.tif
101fc84c9e0758817294b27d1a55fbb9
ba7a85c74604461b648c8072abf77121f3ac5d7c
5363 F20101206_AACVZM hajtas_m_Page_010thm.jpg
2fa0282797cc458179470938ce0f8f6f
382be90abffa7cae499a80efdcf61602c91ab3de
5848 F20101206_AACVYX hajtas_m_Page_011thm.jpg
a018556cdf300df97125726733330b1e
dca67612ed19a0f541f3ae03b1330cb408d32174
6628 F20101206_AACUWK hajtas_m_Page_013thm.jpg
0f5d37bf3b169256540175ba60d82a5b
a7e06e2d6490711cd527ed4d80f06cdd70896feb
112741 F20101206_AACUVV hajtas_m_Page_048.jp2
dbf817f19104a3923da2ab7c8857cf55
1615eb707c7f6514d99def43257ae8b03883205f
6873 F20101206_AACVCE hajtas_m_Page_108thm.jpg
9bba72bbad4dca19f0bae94b79cb77f8
75abbdd7f518a54e334585670351c7f947b2787e
54993 F20101206_AACVBQ hajtas_m_Page_110.pro
4d117e10011f95c7036e32cd55e28a84
1cfdda2d446ec711b43d909dc04f396fbce68b5a
6943 F20101206_AACVZN hajtas_m_Page_015thm.jpg
57690d4fadf0bb60febcdb0f6f39b743
edbc978ff17adbeda52c786ad92898282177b9ce
16912 F20101206_AACVYY hajtas_m_Page_005.QC.jpg
9813be4dc92526af0fc84c01f5ea37d0
3ed7c354c0c72082d0ed3403ca70361d84301282
86595 F20101206_AACUVW hajtas_m_Page_065.jpg
9e954252cb48980a6613006340e77a85
65354f5195adfc38d9b9ed3f5a23d6ae61649ba9
374 F20101206_AACVCF hajtas_m_Page_008.txt
d12a8a7867b50c39bce3e0f06bda9934
f2bf4301df5302ec198a43a144d6140f4c5652ef
1051908 F20101206_AACVBR hajtas_m_Page_049.jp2
0a5906ec462d4e2799a0f9305975b69c
384cb0a7dc6e3c7fa3ca0dd14978c38704803a4d
550 F20101206_AACUWL hajtas_m_Page_003thm.jpg
6c82d0a5b230fe987274e5ccc518e25f
70feb676811ea7d37513544fe74570d55271b946
30345 F20101206_AACVZO hajtas_m_Page_015.QC.jpg
a56c5856b349c2d9e3626bd1f529be4d
b4d699cce90a8b2635505cb29ad666d304fa05c3
28517 F20101206_AACVYZ hajtas_m_Page_101.QC.jpg
23ee8a4f8156e36367a2229d13bc73ee
65b2023834e1ffb964ba4866f616854fc77d7763
133923 F20101206_AACUVX hajtas_m_Page_055.jp2
f8e232ae670d266782caa1a9632120f2
3bbffa7ccb213043538391ccefb41d152c0bbbf8
F20101206_AACVCG hajtas_m_Page_045.tif
151ccc8e5a3fbcf76e2ea935b83556c8
dd7a18cc34274c1588d760b8a8900536b1c26323
F20101206_AACUXA hajtas_m_Page_043.tif
2778cd2f38621e64b57a3d582948e013
d6cc2c6d51dae6435d71db81722026f85e4551ed
86211 F20101206_AACVBS hajtas_m_Page_007.pro
75a5bff06e710027dd16cb1965848580
db6ad78aed8c915bd2c21aef2bb4ee66dc1815b8
80137 F20101206_AACUWM hajtas_m_Page_006.pro
6b5b6297f8631f4326e7ba0a7ed02627
3e0c10e4b8b4a64b15f7ebcd320ba30dc60e9680
6978 F20101206_AACVZP hajtas_m_Page_017thm.jpg
1fd0476760b88520f437bf03408a177e
d592dc6178e3303ed27020c0fbf8a5e237b07dd0
94271 F20101206_AACUVY hajtas_m_Page_080.jpg
066f282aff75211bf81f81e52fb29002
59fe9dc1de0f62da541c70f958ed7e5fd2d6d2cb
28063 F20101206_AACVCH hajtas_m_Page_066.QC.jpg
e0ffd693c4a8f731b315992ee99fadea
a49a2e6326eea2ca3de18fe694f3b3b93d5df635
6726 F20101206_AACUXB hajtas_m_Page_037thm.jpg
78a37e8d65f22a636f464dd6c5345592
caf7577fbd09ab3b3fc4a55d53ed37df5fbd9b83
29089 F20101206_AACVBT hajtas_m_Page_097.QC.jpg
cfde782a4ce9c38849c1c38675d77b3d
a22e764c344ab88311bd5de1537643167623d45b
F20101206_AACUWN hajtas_m_Page_042.tif
ebe2cb5e5273b2d4080969be20d40e01
1dc58030582e314664dc3acf87f2ec95639ae92b
6570 F20101206_AACVZQ hajtas_m_Page_018thm.jpg
60d095aa86b532b904d87d9a81ede715
357a537195b8d4e3b1e309f1ed236533b02bd435
74369 F20101206_AACUVZ hajtas_m_Page_138.jpg
5723043e6c9bde319b76e1e756d405e4
d39b078006d472038cffb07ca571de18278bb9ea
54087 F20101206_AACVCI hajtas_m_Page_094.pro
42c903781223eaab34545b2bb610a340
1c762514337204b34238a2061be2eb49d769b7d9
33274 F20101206_AACUXC hajtas_m_Page_064.jpg
f28acf6795b57bab0e464b5793b420a7
cc68e1cd11cf27cef8a7a9c1120168238b2aa3ae
F20101206_AACVBU hajtas_m_Page_021.tif
d8cfa01718c933aa0f560dbbc49ea50d
234dc89452ddecaf398c7838b2327699c0703760
8423998 F20101206_AACUWO hajtas_m_Page_088.tif
1028c06bc1cf2fd3d025d77afe02bf54
52b5d7bbdc1835ed626c64d1fd91abec65bee465
86747 F20101206_AACVCJ hajtas_m_Page_071.jpg
df8d5ed4b0d3749b290a4e103aaa2bf9
3109256a89e35d902e7984ff0e67f13fdf1cadf1
92692 F20101206_AACUXD hajtas_m_Page_107.jpg
17b818aeb5334e8658c36519be5e44c5
ca6a1f0dfcb3cf042b31ecca810e249ea053ce80
55964 F20101206_AACVBV hajtas_m_Page_131.pro
f035987e7a1c42a296e19d99b81680f5
ea1b1baa4c12ece939becf36728036fb0d942b91
6886 F20101206_AACVZR hajtas_m_Page_022thm.jpg
e8a7098c053b76213ad5255be1f4c3de
626f55a241fda9ff39ba859f64f0f6e42338a7b5
F20101206_AACVCK hajtas_m_Page_032.tif
55341b080c090f12a54cb74011bedb3f
5c113eaefebd4ba3206addab11e38ab9ccb93051
F20101206_AACUXE hajtas_m_Page_069.tif
9be49bd517202bcdde1ec4cd58cc9525
cd163e0c1c5811ddc9fd69763512477e9c63360e
1319 F20101206_AACVBW hajtas_m_Page_039.txt
80ebc79d4e923446e037f19668d10933
1e6ed425e5dd8869da6ee17ce86f9db853225120
2202 F20101206_AACUWP hajtas_m_Page_010.txt
e4d99a23626b292b1c9b2405adf900c0
e78d726921bd01763acd47e0766be8770cef9126
28990 F20101206_AACVZS hajtas_m_Page_022.QC.jpg
69f22ad8a659a1688ac63d478eacbadc
afc7f214491adb3bf3df6c02a36f97c455332784
80608 F20101206_AACVCL hajtas_m_Page_128.jpg
774db93ccbf7acfa79dd4881789f52a9
831fb3d0266b82a0921c30a71d1f9e7e948d6926
30061 F20101206_AACUXF hajtas_m_Page_024.QC.jpg
0f0828269e4ba64e046f9046a3c485e2
62346ea667423920cce015e7ea13f2dca9e3a52e
14777 F20101206_AACVBX hajtas_m_Page_087.QC.jpg
05138a53db9c46d2d355f632052ff13e
d1adc9f69f5685e6d98cc5ca2ac3d593d020df54
20301 F20101206_AACUWQ hajtas_m_Page_125.QC.jpg
13521a52ab44f56febf19c83d05df1c6
0777aa21164fe30688554d26b87daa6265d1bde5
6884 F20101206_AACVZT hajtas_m_Page_024thm.jpg
b0bbbd0cefad633e967e62a068c7e8df
ea5c8eee04eac32d95d0552f9ba811ce68aff722
F20101206_AACVDA hajtas_m_Page_015.tif
7539a778f76681f7548f0951c664c6dd
3cb393a66c55d22e5af352c9083f98125b715080
114597 F20101206_AACVCM hajtas_m_Page_072.jp2
1673ff668608a2d945b5eb126741c489
267d28c26b68134cedc2523384027463a6150cf2
5398 F20101206_AACUXG hajtas_m_Page_007thm.jpg
d06b7bc9121ee44ecd77ced9dca5e318
98c08100e30cdbbe5b5f771393a680b541221e77
118600 F20101206_AACVBY hajtas_m_Page_032.jp2
fa6af5fb48aa852048028ca76ffdce3a
e8001aace4ae63278a88e522be73731cbed07f69
6828 F20101206_AACUWR hajtas_m_Page_020thm.jpg
65e8ee308f356731ced1f451b9299080
520a43b61b30affdf54d454d05bb85958b5be6eb
28581 F20101206_AACVZU hajtas_m_Page_026.QC.jpg
7a55673667ca5658914a05963e75af37
1d0c553e058e387c8233428204879e6ccbfb1631
5642 F20101206_AACVDB hajtas_m_Page_131thm.jpg
2a59c7780a1e27b72cef502ecd5d4684
81bd539387319d002f4f548b96bb07065e920969
F20101206_AACVCN hajtas_m_Page_098.tif
b4ac2b5ea78bceecc97f1b485805eb90
ad3598a4c2940a5e908186b863cc23c0f30e5578
90329 F20101206_AACUXH hajtas_m_Page_050.jpg
4e8f75bf07c5073cd5cf97cb428e6903
b8939fc552856171ddef96c5210fc25c6bb47c54
459 F20101206_AACUWS hajtas_m_Page_119.txt
72e1e29afe9d3f3b02fb234f05f9a02d
68f2412160dac7d73eb4a7968648c9875616cd31
26856 F20101206_AACVZV hajtas_m_Page_028.QC.jpg
426891caf4895e09308c6a4b40003414
2802f663b46333ff3c2ab5bd61ec226889bc0466
51334 F20101206_AACVDC hajtas_m_Page_062.jpg
ce927c8295dfe92c121ea8d0e2ae4eb9
1b8ef541f05302e0698dabf69445136ac3ed515a
6893 F20101206_AACVCO hajtas_m_Page_016thm.jpg
aa158790b10efdd687b05cacbf72433d
8b6d68ff5db92990fa1550dab91512b1f2cbfe32
29116 F20101206_AACUXI hajtas_m_Page_093.QC.jpg
b1a3b795f2162a0e9449499528014edb
dc69b86ca250ff89d4b43f7ef952988aa9aa7152
1051934 F20101206_AACVBZ hajtas_m_Page_024.jp2
6ac21a6691e889be9412e702db993591
57875bc726cab2469521d74a4063ad353e05eb95
4026 F20101206_AACUWT hajtas_m_Page_005thm.jpg
684d774cb707239816b1bb8296f771a1
2f6516e20c46dba7a03b7508f9d0790447aabf0a
6761 F20101206_AACVZW hajtas_m_Page_032thm.jpg
33200836ea89281b8094e338c434e164
1a8f21a9a60670cefaf0a373e9ea8702930b4f37
5067 F20101206_AACVDD hajtas_m_Page_123thm.jpg
f2af56d63b6af5ed94534426f7680d30
eb99cd2a3b6e90fe3c397a01cf365e76204b15b1
120863 F20101206_AACVCP hajtas_m_Page_037.jp2
9e9870e0c6fedc4d79d5da4bb41da5d6
276125e848391b2a5e8808619c053bc9858a6640
53218 F20101206_AACUXJ hajtas_m_Page_010.pro
a830f9255809bb66a34ea390e7149bec
a5dd22734d1672c9c1cd1596754ca3ecd859c921
28504 F20101206_AACUWU hajtas_m_Page_074.QC.jpg
bd86cbbbe512200656cb3f4724ecf72b
f19ad9d49b072b454cff8fbcf6c603709ecfff0a
28604 F20101206_AACVZX hajtas_m_Page_032.QC.jpg
77bdcca2d7fdd4d99fab3cceb07c2de1
bcc2313e2bee065827327b64ecb0a7d42ad3cb9b
25839 F20101206_AACVDE hajtas_m_Page_057.QC.jpg
50141e4786ec80513394a5448fe63d28
a6baf225713f9e467ebf45f27f603c0056b48578
59288 F20101206_AACVCQ hajtas_m_Page_015.pro
15e867f9470da6a774633218873f3123
096ac5feb0f0be794e5c9732026b3e26c1035ff2
58143 F20101206_AACUXK hajtas_m_Page_017.pro
dba81fbc80b72db3698db92cb1ba5cff
5338153214a1c0371c6ecc8a6c052f6be8455873
27784 F20101206_AACUWV hajtas_m_Page_053.QC.jpg
251f59de0abd1b3614eb6056f4090b86
50bc331dbf215da030038a6878cfed375ba81c0d
6498 F20101206_AACVZY hajtas_m_Page_035thm.jpg
f67f9a5f6aef87b7163ad369ca4a55e2
8cd5d7173103790b8f419400ef510d7cc0456609
1488 F20101206_AACVDF hajtas_m_Page_064.txt
ae4da3e30e86084843ed0bc0be5aab05
754bf54c4c7ec282ede015454cd0c4ec399e9b29
2056 F20101206_AACVCR hajtas_m_Page_066.txt
60197462b98db8885b2d8b3df2ff8dfc
dc331418c28a47783a3dc5c1253c204e0e93becc
117930 F20101206_AACUXL hajtas_m_Page_094.jp2
54c1fe40063e7cf6530a2a6bdf95fa0d
2815f8f69680ba1c22fc9e9666e24403ded1d96a
4857 F20101206_AACUWW hajtas_m_Page_122thm.jpg
4245515671dcc8100634e09ac40a5c22
9fc02f8ea25fef9ff572855bc4c2b6a4b9f2e4e1
6122 F20101206_AACVZZ hajtas_m_Page_043thm.jpg
73f3415817d47d7af989cc752c48f205
243b8d4e519b6b9fb32418629e193672a39b3773
117183 F20101206_AACVDG hajtas_m_Page_020.jp2
e85533abb8fb948b6b8e630478950bd2
4c82ebc5f60765757c559fdbac0d62273e305530
107848 F20101206_AACUYA hajtas_m_Page_084.jp2
6df714bc30e15393fa3bdb424de92455
2158fddc11d04345c1aa2d5359ce105a5527f1a4
27852 F20101206_AACVCS hajtas_m_Page_068.QC.jpg
05497327261f484030942ce36f139eb3
54f6ed4f93b0e403875b4c0e89919465e04f545a
57732 F20101206_AACUXM hajtas_m_Page_059.pro
6ee5a4202e60c0debdde2b95e30b61c0
ce9a0a08c411b06f3f3a1fcf1a03ffc945f03cfa
88076 F20101206_AACUWX hajtas_m_Page_018.jpg
02c42795e7329328e541a44f9c7980f2
672ebf8a3803798b178e4977c206826038605863
F20101206_AACVDH hajtas_m_Page_034.tif
4485de44a1be1991d115730e5c08c09c
65043cb2b7d391eae4c4e83527be931403429214
4493 F20101206_AACUYB hajtas_m_Page_118thm.jpg
9c48846b8c981946f228acebb3796f6d
c7adf41a803adcaaf7e3a25685897e7877661e8c
56105 F20101206_AACVCT hajtas_m_Page_052.pro
521c1c82fdde7fc09817187ceb44defd
e881d0f08276c8af0ead794bbfebdc4d6fbec826
2680 F20101206_AACUXN hajtas_m_Page_038.txt
29311286d693a99c7f5074b3d76479b7
1969d1c55232fff17894d5e9c8a9d3050c2a3963
27379 F20101206_AACUWY hajtas_m_Page_077.QC.jpg
bae401eee2e5d39502fd6ed7181e133a
4164ec6f6f4d439d7d7b253342bb8243ce18c042
55528 F20101206_AACUYC hajtas_m_Page_032.pro
36ffe2a2dd63712a363824a13c8f7fe3
d7cc550877d6a83b8f17efac57e2780e61ac4f7b
6568 F20101206_AACVCU hajtas_m_Page_094thm.jpg
7f5f22c7864dcb60bd711aa12ce0690b
94b156d34dee9520df236428484f4b66de511dd2
2102 F20101206_AACUXO hajtas_m_Page_036.txt
6c798cafbd6e0007396650f59ef96aaa
66b07f18367b008146780b51849fec9f0c97ed9f
F20101206_AACUWZ hajtas_m_Page_060.tif
ce54521f7bc8cda784094becfa40e700
ad6cb88cec05efb98b7f7968a0613abb756eda4d
1051973 F20101206_AACVDI hajtas_m_Page_059.jp2
98ac0d68c24e2a784158c10527e553de
bd682938f58852231747e2a43f102aa0f378236f
6637 F20101206_AACUYD hajtas_m_Page_058thm.jpg
6d9bf118e0f215c02b032df5d356376b
a3fcba49c0b1e888d532909c7efae9de91f7d3d1
F20101206_AACVCV hajtas_m_Page_055.tif
1a4c29047a1244c7760eafe3dc75c1fb
b75c548fb3e5076640c3a5e3f2ee316fd0164a81
17577 F20101206_AACUXP hajtas_m_Page_121.QC.jpg
d7599d5844221b8d9cd4e80c35db7ba5
75422ea8f53cd2a759005b85a21ef87b448ddbb5
91050 F20101206_AACVDJ hajtas_m_Page_109.jpg
3b200a2dd1ba4c06fc17c06de763e37c
2c1394595ca0cdc004f3477d1182840fa581336f
29446 F20101206_AACUYE hajtas_m_Page_099.QC.jpg
3cc4d91360cc9d7e804519e6e864394e
3ec0e62d7d3dfe019099ea18e5206b363a490052
89576 F20101206_AACVCW hajtas_m_Page_036.jpg
67eea978599600bd46aa545e534713b2
d3314d5dc94f148ebaefe00e1c85906fff1b441b
4820 F20101206_AACVDK hajtas_m_Page_137thm.jpg
5ba94196b8508e6e33c4b391733daad8
3d5ddc3e34d2806cea5420ab0abdbfe1332f2607
1577 F20101206_AACUYF hajtas_m_Page_086thm.jpg
cfe1bfdbe883d3449fe2b5f9a8d36d3b
1382e7ffefabb106d8d934727c0d682b43bd0c2c
28561 F20101206_AACVCX hajtas_m_Page_027.QC.jpg
c847d1d75274012b94f6b043c737daaf
e994567b42676b9af446159a741160f8048663bd
F20101206_AACUXQ hajtas_m_Page_044.tif
4e571b4e82722fea02414f9a6a0377e3
952d67af1cf8432e92c12ddea7227945d51a6a83
30450 F20101206_AACVDL hajtas_m_Page_063.jpg
529be18a81826c71b766030858f564d3
0bc7e7935155c7e34edaad813cc775bcf2eca02e
6788 F20101206_AACUYG hajtas_m_Page_026thm.jpg
47f3dbc290716a4ee888c21f86286ebd
0cdc65fb244aa664e24fb6b2fcec8af7080073f0
2026 F20101206_AACVCY hajtas_m_Page_049.txt
716f7263787b3bead53a2fad499a5721
5982ddf3e5c32e3464d5fcc8b9d4391a42473c96
6499 F20101206_AACUXR hajtas_m_Page_095thm.jpg
204b54d1ce1370815bc78bbda38be6a0
6d54f6f5c577d3f8cd12576ff935fdb48d734bf7
27446 F20101206_AACVEA hajtas_m_Page_072.QC.jpg
9f0ed90f5b08f11c14f1b1db51bfb7de
9d7b479c7259a4e884f955869013363e63d3fc7f
4684 F20101206_AACVDM hajtas_m_Page_126thm.jpg
2b0a09a62414c394cf0e706d60a9447c
bea344a3ede399c9493658efd37b452a05d5a024
2141 F20101206_AACUYH hajtas_m_Page_074.txt
f94826ce85ad23f84644633dbc90beea
da9d46fb0be1bc79613eb371fe70af492b3d1373
28787 F20101206_AACVCZ hajtas_m_Page_107.QC.jpg
aca209c5b6a0d3c35f3fbe41e0d2bf06
7c1c757d61e6dc5b396265bd88cf7ddaf0bc0a75
6720 F20101206_AACUXS hajtas_m_Page_053thm.jpg
ebe6ac3d02d2bef2533af229e753ff7e
17ae30643825d439851adfe74f19622022a3db5f
87469 F20101206_AACVEB hajtas_m_Page_068.jpg
ccd080ec7f3d832c3da0638635a6687b
17f12840bbfe078a728c788e57727fe8d9fab7f5
7099 F20101206_AACVDN hajtas_m_Page_106thm.jpg
04557c5c9b2f5e9fe425e42519f83761
c70105e36838f9b7af80825e0893453ca23a3f92
F20101206_AACUYI hajtas_m_Page_136.tif
efcbd8835590583c84b124d3c9d3b0bd
142562e0b3daf540546feab6ff3721c492610ee9
91691 F20101206_AACUXT hajtas_m_Page_022.jpg
69a2716db0469ad043b36071b334024a
4a45c6ae094ae434112aeefa3c8b7ac5f28f940a
93197 F20101206_AACVEC hajtas_m_Page_078.jpg
ff2e8b5e3ec300366a19f6166df4afc4
7f8336146f0383e3c03f20f5dfd43cc8ae74ade1
2121 F20101206_AACVDO hajtas_m_Page_053.txt
600dc791450dc5f48dc93e6f847eef06
c3777f3fb05829a9a699427856e4a77e651cd1b3
52890 F20101206_AACUYJ hajtas_m_Page_005.jpg
a567abf59fca4ef0f8d15d673b477a96
ba786d3dfc2c7019393e62b9264384a39c8cec9a
F20101206_AACUXU hajtas_m_Page_125.tif
c0a82fba3712a35c2802e3556ba3dba4
e36a61c2b7c0e16c1942cf05b2b089529af72b2c
30284 F20101206_AACVED hajtas_m_Page_031.QC.jpg
335149031700595304462fc8bd800ca1
fe7532cc38ea5a9031daa2fab69fc343125459d8
46360 F20101206_AACVDP hajtas_m_Page_043.pro
d6cf9db4164341339793eeeb495cf3d5
f430cca4d4164c0ebfa65f459355c673f270a1cb
28434 F20101206_AACUYK hajtas_m_Page_096.QC.jpg
b6497181b3f49e6836fd6ec65be6c368
805d7980b261c8557b9dc219b9d0d14681da0e5c
F20101206_AACUXV hajtas_m_Page_105.tif
8d16bd6a6447aed99f62455fc593dc4f
4e5e8ec40f45c60c81523098e4c8c4bec640c023
6778 F20101206_AACVEE hajtas_m_Page_074thm.jpg
09226f2cdca3f9b7044c7e6a7f85c3af
413f59578077952b96f4e04e10269f7861f7eeea
F20101206_AACVDQ hajtas_m_Page_119.tif
7b2a6a8fe778c9890182c52311d0b675
aabb04e736411f72b8f7661c1e2acf66a340f5ff
6988 F20101206_AACUYL hajtas_m_Page_023thm.jpg
3ffdbdb2f2cd880f33e70b46a6adb6f0
fb2fd94e53c60b3822f09825592cfa1654d84420
114361 F20101206_AACUXW hajtas_m_Page_077.jp2
f430ec4ffe5c351e7e0ce4097db37bcf
cd5e6ca47c7b783b6e27cd66c3f3e23b9835bba8
2152 F20101206_AACVEF hajtas_m_Page_073.txt
ac1f3c6b97f8445776f491ee5ae74758
b14e1383d4f6cdb40099378315c9c30408d09ba1
120639 F20101206_AACVDR hajtas_m_Page_108.jp2
a01ab2bf3013c90f02a91166b331a737
21d3fa45232c07c20c17d5e336dc9666205e420d
6694 F20101206_AACUYM hajtas_m_Page_083thm.jpg
efc08e5d9eb32d29ca65780346b3ad2f
8d5130c37a9a6cdf70cde24643ee2e0ff55e4717
F20101206_AACUXX hajtas_m_Page_101thm.jpg
ef10ea99cbdb6169ae20c00752ff715f
2c037fbb7df589c70de4dd192a3eb07c8d02ad1f
2074 F20101206_AACVEG hajtas_m_Page_014.txt
d7621cf129f812dfbcc26f341238b9f6
654e946c0eba98484ed68e08aaaf2d9adc6e2c73
87343 F20101206_AACUZA hajtas_m_Page_054.jpg
d73a56c1312d085c4232f1ea7fae15d7
0cfeb7d74d3dce5433304cb29707843fa38f238b
4722 F20101206_AACVDS hajtas_m_Page_117thm.jpg
4d42f5830a8b5dab4255405b07edd551
7df373f75a6400c363d90a8f0ba85fc319720631
29565 F20101206_AACUYN hajtas_m_Page_098.QC.jpg
f58ae339f32585a6b1aef01a61762f47
ed7bd923d58595136d319b2796b645dbe0c223f3
88822 F20101206_AACUXY hajtas_m_Page_020.jpg
be706731c6e647f24c50554f631a4592
7e67afda5c8097fd6b4891a691f1fa5474679d7b
F20101206_AACVEH hajtas_m_Page_070.tif
f79014484c134324c39c224fd91c318f
e7bac5c139748f5aa5bf82f4a77f8a7ca3a852d3
4584 F20101206_AACUZB hajtas_m_Page_114thm.jpg
bdcbfdb1e73d56c1ec6c710965c8e192
e0aa3ed152527b3613033b429cd5eefc35f1fd0e
54499 F20101206_AACVDT hajtas_m_Page_073.pro
d9cdd071e433c28f2b6f9309821401e8
769fc71affe26a527c35404fda0a337c34a3f7ef
1051936 F20101206_AACUYO hajtas_m_Page_036.jp2
be1bafd4f986a54ef56d6cb81d6dd654
01bced968a0f0e04706034bbc4eea6106a752c5e
54510 F20101206_AACUXZ hajtas_m_Page_044.pro
7338ecc6c1fd6e3074f5acd12d15ba47
6e445b3e19a290818754eb4466bce00e21be582e
29203 F20101206_AACVEI hajtas_m_Page_056.QC.jpg
067e1cd04635d2aa782d3c8b7af89cfb
9234fcf76c7b5e8e97e2363ef89fb75bb5b57c10
52526 F20101206_AACUZC hajtas_m_Page_035.pro
de59564d3e004b32db426ce368b10013
1f579e13c64d083d32746abb2ffd11a077451b8f
52255 F20101206_AACVDU hajtas_m_Page_066.pro
718515760a0abfbcb23632784af606a1
09630f44761e4729fed1a1eeb63853f93d90604e
4877 F20101206_AACUYP hajtas_m_Page_124thm.jpg
a1f03a7fda9d762f8d068178124ebe8f
145784e5d54b96d03312b80b7ef5a0777fc36b70
58315 F20101206_AACVEJ hajtas_m_Page_104.pro
3bf25a50d975482521c71163099c92e0
fc0243968b46020feeebc9a90314030318dace99
F20101206_AACUZD hajtas_m_Page_008.tif
acb40b8e228785f4bab858c8f0295d96
c7a4d5b1db38d59e08b54406357cfa93cc07052f
27540 F20101206_AACVDV hajtas_m_Page_046.QC.jpg
8f71ae667cc587c7e561f1b86555d206
1b02a0e69344b1068cfae13b9a519d6214e503cb
23242 F20101206_AACUYQ hajtas_m_Page_138.QC.jpg
17cf60819ea671f311e2485ab5e51bb0
97c17ef0ee1f625e7274eb5a693b713eee4daa9e
26560 F20101206_AACVEK hajtas_m_Page_115.pro
223f64f2cd8487568d6da800ae65199a
e7b646b6dc57941c321542f9aeb721833aef1018
29376 F20101206_AACUZE hajtas_m_Page_023.QC.jpg
f45ec7f1c50ab3fafc853f84084dc0f8
b6fc596d136a31094c13b43642ea47c82ef22853
87422 F20101206_AACVDW hajtas_m_Page_102.jpg
da9ed7a9c0257459d70b84a6ef715da8
9a259480c441ef04fb788f666e450f386deb8ada
1051966 F20101206_AACVEL hajtas_m_Page_044.jp2
df1027e3a29e44f6db9f49e4ef9f7cb5
3cab992a68e6d71f38cb0d498a52db703eb584f4
53499 F20101206_AACUZF hajtas_m_Page_018.pro
7e4517450f85242d441391fc0e1b626c
90398d1ebb9c9362e156f31ffb6a320850494dc7
82037 F20101206_AACVDX hajtas_m_Page_069.jpg
8d91a222afab8aa1424ce5ff709da1a9
c048b7260684cef69c148cceeb25200930dd64aa
85051 F20101206_AACUYR hajtas_m_Page_075.jpg
49c49bac5b72d9d3595f212a79743c76
2bb9bf8f58c200e36ce7a71e8723a99389417325
57650 F20101206_AACVFA hajtas_m_Page_097.pro
34660bb89820d87b7b194c41185f4990
2057a87dd7fc133a5c9e2134916dd65b36f57eef
24172 F20101206_AACVEM hajtas_m_Page_012.pro
4f7f09d441be71542efd628cd49575dc
38501522e15533e9b57a4017246feb56a69555d4
17764 F20101206_AACUZG hajtas_m_Page_116.QC.jpg
4ad23571714da9b3615c28bf917a9be8
e5f72c22ff0781ea6e8df726b0c7cbe2240f686c
46551 F20101206_AACVDY hajtas_m_Page_127.pro
ef8c41c94bd517f76fc0bd0878c0f594
06213bc6c360035b5c5dcd8284fbb29fb82237db
124633 F20101206_AACUYS hajtas_m_Page_111.jp2
4b3ad80f9e37ab3355a22207f7d21708
1953143cb915a0e56eb29bf42ce49aeffb6edff9
4638 F20101206_AACVFB hajtas_m_Page_003.jpg
8e09863cf644652b020e3807eec6e538
95bfd1a6465e28b1d8e5dad05b213dd0704d788b
2048 F20101206_AACVEN hajtas_m_Page_004.txt
22fbe530c0759c575a3618fb7a1bdcb1
bbbcc95311594def85247f97341e49f7d6902288
124608 F20101206_AACUZH hajtas_m_Page_017.jp2
0bb7927e5ef620fc8743f69060d24ecc
115dc6f3e0ab14c1b91af903c9a6192e78371966
27903 F20101206_AACVDZ hajtas_m_Page_013.QC.jpg
d94a60625a0bba5daf68629f24b8334b
a657314480202ffef1d5c212bb8884e40a711232
2081 F20101206_AACUYT hajtas_m_Page_092.txt
34dee85a52f7a46b18e9d820e8a82048
50e800e1ab48c3901be62dc46ed674dce7db8385
2134 F20101206_AACVFC hajtas_m_Page_095.txt
6c09846d3e0774a5eb90a01989b0d34a
94a1240afb86e785a94b6c97c2f5355d02a9016b
6541 F20101206_AACVEO hajtas_m_Page_030thm.jpg
4fc35c8a20068742cb726b1cf27aabd4
a62e08f5542a61132c5e4895f1256045bb809152
120386 F20101206_AACUZI hajtas_m_Page_130.jp2
c73af99b39d1bce6e3fa081d868ed02e
777c5e809f67a8aa79f50b8d1f6700d064b1ded0
50976 F20101206_AACUYU hajtas_m_Page_060.pro
4de97b5044b6719596f949a965768c7b
30bc30d077829a7f4bcb29168ab5d08da38ec04d
F20101206_AACVFD hajtas_m_Page_066.tif
9dc40e615b9d350ba82772c2ef65af2e
904c6403c2e263ce096684f004c85d47e88b4f5a
6096 F20101206_AACVEP hajtas_m_Page_084thm.jpg
f7bde4c8e29fdde46dc6b27e131c0665
c8e13296209cf8855fdefec047f455e5ad73e71b
23451 F20101206_AACUZJ hajtas_m_Page_007.QC.jpg
0d4ca76b99d50e8a4bad8022bd0f9339
22371ef241f4cc41615350cc897c9c05cb46535c
56029 F20101206_AACUYV hajtas_m_Page_101.pro
3002c6b8963febc07d653679ccb607bf
fb74374f05fb7a0c9fa272b6a4c2159234869c75
1051959 F20101206_AACVFE hajtas_m_Page_026.jp2
586701a96ed67ac2654f195cbdce6873
d11bba05c9564e41a74ccd85c2ebabacb7f9cd03
94505 F20101206_AACVEQ hajtas_m_Page_098.jpg
699c01305a2e9430b590e6abc5af2f7b
5c47e436e22220f27441b03b4de21af12cc46935
9884 F20101206_AACUZK hajtas_m_Page_089.pro
079bc88bd768e8534e11ff5b6dbfd2d0
1a5baade639adb87b77537b7cdb23f153ae05462
139667 F20101206_AACUYW hajtas_m_Page_136.jp2
5f34bba2b3834ca520db63e26ffcd564
494031e588f62edbfff08fa9742302b25438adb5
F20101206_AACVFF hajtas_m_Page_085.tif
08b6d6081d9d4231f8965b7d38c3611c
75db98887b437fecfb0252d06a1dce02c0198760
52318 F20101206_AACVER hajtas_m_Page_068.pro
c8b977ea85d33c0601c027ed2b89d64d
61e898656ac5e3ebb04b7916d767445f336d3877
93541 F20101206_AACUZL hajtas_m_Page_021.jpg
d6cf8b0f1f7bf77598f3207efc85fe2f
bedb4a9723ec924582b44255eb4e1d286f47fbdd
1837 F20101206_AACUYX hajtas_m_Page_043.txt
d66d59e20dac913ef41adb38aac36bb4
e87f94bf0172fdbdf8099264f1920189913aede4
2625 F20101206_AACVFG hajtas_m_Page_136.txt
02a9829c0bb858c9ffc02f5ecb19b742
474ba426665b1098c8490381603315784b73049b
2169 F20101206_AACVES hajtas_m_Page_080.txt
e55bede30df89d2672587d3f8681e20b
60972e663b1595e56bc3f561cb24aaf109f678ef
1451 F20101206_AACUZM hajtas_m_Page_085.txt
388b8741cab2c28ecfafffe8dde8c6d5
4f2c74ef5257d64ed718120227d43e4b6cb9775a
113632 F20101206_AACUYY hajtas_m_Page_028.jp2
9c8f8b12cfd4247571051978b427ff07
6b97e99b023e0dc26aa75d2ff72b692cbc36b973
F20101206_AACVFH hajtas_m_Page_063.tif
8ffcba0136435b2702cf67a2b36fca67
8087d8a4fa40a3b1a6fdfdf7bd0128f3b4331095
29903 F20101206_AACVET hajtas_m_Page_038.QC.jpg
8d06aa40859033a9953b730a5a49f039
62e8a0a4a550ba8eff81a14df2f462fd02c2b6a8
F20101206_AACUZN hajtas_m_Page_012.tif
a0f416c138e863da8f3abdead28d4472
c73f95f77fd5ed1ce8e36fce217e9ef9fcf39a6d
F20101206_AACUYZ hajtas_m_Page_081.tif
826e7c814c9c408181cd88d710830b06
ba1367d7c2de86f76b4332b409936bde3abc7851
F20101206_AACVFI hajtas_m_Page_074.tif
c866c206227e0e79fe190e2bcb708f77
efc51ceb1051cc7ec3299f6c40bf85bf7b496a13
118806 F20101206_AACVEU hajtas_m_Page_052.jp2
3cfca202a338277ba206e0899bdf648c
c0e068c5b4612650814c01cfa6fa199eecb0ab02
F20101206_AACUZO hajtas_m_Page_080.tif
51212c7d33439f7a902150efbe12514f
d120a79c6481eed670e8105ae3f389c151530a44
6432 F20101206_AACVFJ hajtas_m_Page_041thm.jpg
aa22925e37d59ad455c0650f15e2aec0
73f024801cecc1167bf5214cf113d37f4ace2df2
841 F20101206_AACVEV hajtas_m_Page_061.txt
c23ba5e9555dce61b5a0100d5295803f
bae86578aebdf00e934fcb98331ba4262b59cc80
27389 F20101206_AACUZP hajtas_m_Page_018.QC.jpg
38febce8ad80c51d0352446e0b5bb780
42e5018863a9ca43ce319abbf8e50b933035f0bd
56986 F20101206_AACVFK hajtas_m_Page_093.pro
bff8036dc4552b88f57bdbfe650dafed
421435b00f4dbc0dc3ff4cf2167c0f989cb46337
28293 F20101206_AACVEW hajtas_m_Page_036.QC.jpg
f0dd52539106595c1b892e14c95652e9
7ce2aa793468d2315a145a25f390a0b44d18daef
F20101206_AACUZQ hajtas_m_Page_131.tif
0758c3a26ca0e1fee9ab860c6a2b7217
33157d95727942a084b96c36b7a069eb2b82413c
89261 F20101206_AACVFL hajtas_m_Page_100.jpg
804cc5b24129b1d13bbe624de5493acc
3262f1e448f83fa223f2320442b52983950b9cd7
58500 F20101206_AACVEX hajtas_m_Page_111.pro
b92cef84dbbaf309505cc2ac79eed522
43fef7d08a563c023884792aff120875a62ec876
81622 F20101206_AACUZR hajtas_m_Page_045.jpg
794e7b5a9ec4e84c9966933b2bcf2963
7526e377173f0f5accdc0afcea9adfa547f1c86d
18238 F20101206_AACVFM hajtas_m_Page_117.QC.jpg
1c31ad227e0c98c241ab59731831606c
0a1fde4ac050a648cf6d97fcff07411f47fbf133
54371 F20101206_AACVEY hajtas_m_Page_095.pro
fa4b6b697b20a6a537b632b783cc12c7
e39d902b086988076a882d40c4be087bc5016c28
98846 F20101206_AACVGA hajtas_m_Page_006.jpg
a8abbd9de423b435b1ded52c557398b3
909c699e1b63dc2d39f339a6c71997fd4dc67231
14579 F20101206_AACVFN hajtas_m_Page_112.pro
40240815af632ca529e3b3e40a8962fd
ba9755d8c8127f69d0f7677b1fe0b475ad1eb015
F20101206_AACVEZ hajtas_m_Page_120.tif
026c6d7cd73ab8ee3a0f6811aedef524
6ec169868ecce469ff2b7455df162c3f3d3cb7a0
55408 F20101206_AACUZS hajtas_m_Page_030.pro
3647af62369422639a3fc3b638762a37
003361c55438a08876913ed96979e14e84278214
108420 F20101206_AACVGB hajtas_m_Page_007.jpg
c9e845a7e2669c7dd7b7bb4cf8d8eafc
fa5d38b60c2bec123ebfa8f7077e999ba19404eb
22575 F20101206_AACVFO hajtas_m_Page_010.QC.jpg
fe9d951b04ec47fe73da56a8efb7b505
599e333a1b026d842fdf0579a9a3fb6e9b2d4c3f
F20101206_AACUZT hajtas_m_Page_126.tif
e0f8cdae203b91fd64f3c54b68cfcda5
fb0658c5ea480ffa8d1cef98286a0a121f1d4fad
10305 F20101206_AACVGC hajtas_m_Page_008.jpg
7fe19691b7abe46d36d726c09331b48d
15c4344dbb09c01b4f6617809e4819dd7a9605af
43088 F20101206_AACVFP hajtas_m_Page_012.jpg
9b9282a8904b31dc5f23bd8ded0ea819
9599337be6e7eb792c4f6c6e932bcdef2725d13c
51621 F20101206_AACUZU hajtas_m_Page_070.pro
0b071fb0b8f0a7cb461c5a09266e74b6
065d4a47edaa7c17fc641c175631718fe4a018bc
12492 F20101206_AACVGD hajtas_m_Page_009.jpg
d0768678ac94ae661bc1403f495d1f20
5cb169a554c3088ce8b00a76d165e0cf593b6fa8
6733 F20101206_AACVFQ hajtas_m_Page_025thm.jpg
9bd0b9d3bacce2139182d6d6d6391160
70883fd12f995a9990ebd86a90c143b65919e183
85282 F20101206_AACUZV hajtas_m_Page_076.jpg
1072f13e865c1f376c8c541d232e0830
146e6c0330de01f355995adc1f2f916b56e97092
80460 F20101206_AACVGE hajtas_m_Page_011.jpg
76f4f27fb21c4afadc969c0f08d38466
a3876a8c8b65541ef6ed7cf9d6e96adee2f130b1
F20101206_AACVFR hajtas_m_Page_132.tif
d3d75095be079312b68580d2b57f42dc
997666440398916772ae35f0cba01fe4378301bf
F20101206_AACUZW hajtas_m_Page_089.tif
9524a0149195c04016085b6bd75e85b1
dc9ac8272cd4633b21872498a9885fdccb84e151
96438 F20101206_AACVGF hajtas_m_Page_015.jpg
9ee2a150f542f537e185760d04f52923
ac12d7d9d657b1b7d2b08297daafe8f4249cdb4d
54000 F20101206_AACVFS hajtas_m_Page_117.jpg
93c26480cc70cefc28430ec3fc194986
6ffacc6ec9053d33e78b727e0884e32d6025681d
77248 F20101206_AACUZX hajtas_m_Page_126.jpg
8c2d642c8bd7d31d685288f9f2c4af0d
f055fb4d336d3e3bd73c8d17a0b5cbac81b4a607
93400 F20101206_AACVGG hajtas_m_Page_016.jpg
4ea0e4abdeca4c0c2711850e34736661
087a5f28e208880e709dda89b4a9368d07a3ccd2
F20101206_AACVFT hajtas_m_Page_006.jp2
f99b623df1389a3f1c17fb97f198624d
5e756b9313e5e1b4e903418c091f0c390866f88e
1973 F20101206_AACUZY hajtas_m_Page_125.txt
f960aa4a7a256af66994068ff14ff302
6b66f0cfb827be211eb26af3955557770ef24544
95076 F20101206_AACVGH hajtas_m_Page_017.jpg
1455ab2fe33956f036a1ac1122d93984
6eb1a454909d32e60031fee74c2846e34f6980e9
26765 F20101206_AACVFU hajtas_m_Page_076.QC.jpg
8fad2f3f96b8744da8a75f7e4297900d
bac54dde6f1230c1b75266fad5d5f9a4fb61caba
2266 F20101206_AACUZZ hajtas_m_Page_106.txt
6b4858f9ddbe6fd9e0ffcf78d562edcc
bd76ec10a5832ad9010f655f10a13316834e044a
90978 F20101206_AACVGI hajtas_m_Page_019.jpg
e071227bc49256f7bece28bcc7f963b3
9c1bc4c7bafe485aa7dda72361d1de8ca4db46e7
159489 F20101206_AACVFV UFE0021429_00001.mets
61851df8306d3d07e1e677b349e0858c
20caa3b3d75ad07d4daab6547aabd962caccdcc6
95279 F20101206_AACVGJ hajtas_m_Page_024.jpg
8d6e269f1c92f70ca8f21f0ab475fa26
3fb2cb2c7ffb29bcb77d62dd017f6f6b1d2dfbc0
92184 F20101206_AACVGK hajtas_m_Page_026.jpg
f41f3c09e01634807008d2deb06e0061
3fd4e00fe55d5840b3c19248c5831518c99c721a
93113 F20101206_AACVGL hajtas_m_Page_027.jpg
9d82e563f09e021e8ee08c3269db2bbd
0eff328acf82817262e625f50227fec27da74585
3606 F20101206_AACVFY hajtas_m_Page_002.jpg
010094aecfa4e9a6d587a3368707a4bc
dcbcd351861c81fbc6dd91ec8b0a384dfd1a4dfc
84648 F20101206_AACVHA hajtas_m_Page_047.jpg
286a4309dfbb4c98cb860418c32da8a6
030a80bb6cdaf67f1f96b17d770e1f543fe86168
86694 F20101206_AACVGM hajtas_m_Page_028.jpg
350914bdc99ca75c9fdae62cd00e412c
5284d279e25946d817670fa8204c54b4335b18ba
86209 F20101206_AACVFZ hajtas_m_Page_004.jpg
2f5512efe8f1e2791f08b50b7b622725
a2536ed1a88e5830b8f48547b54ec3c38ff1de94
87100 F20101206_AACVHB hajtas_m_Page_048.jpg
10b155e8f0de35562f4579acaec4b19a
483b9ad3236849f2034fb03e439da2af11c7df91
91758 F20101206_AACVGN hajtas_m_Page_030.jpg
5eaa849108f63bec98a2f142a3f4258f
6da80720d7e64600e43afb54e348728ceeaf03d2
85515 F20101206_AACVHC hajtas_m_Page_049.jpg
996ef417bf529901501e672240de54d0
74ca7ee701c55973976f7a06fb953dec88791712
95664 F20101206_AACVGO hajtas_m_Page_031.jpg
7e083f701cae4d4b978e7b3e0f4e764c
2a1491b90c5889bee25c176b9483294ace326c0b
91386 F20101206_AACVHD hajtas_m_Page_051.jpg
41e4a5a5349ca00785ee1d86753bda2d
45655fa2d37a2d101471eee81f53c8ea7947d472
92248 F20101206_AACVGP hajtas_m_Page_032.jpg
443cce1a5d536cc78a7a044a43f58d38
05106095917e6b9e8b91d1e71ba5460a1d5bd51c
91678 F20101206_AACVHE hajtas_m_Page_052.jpg
d29231ccca373b11cb0fdd5adefc8d31
ef791665d40a64a464375d8243193f61916a07af
92780 F20101206_AACVGQ hajtas_m_Page_033.jpg
37d642ff82f4298aef265faa490fa13b
bd705eb66fbe320f6073e08d66a55fab6e33b951
89108 F20101206_AACVHF hajtas_m_Page_053.jpg
6d9c22701e33d551b32074a94b599ba5
fa080cec31e116215642d3bccc35a078c0374417
88788 F20101206_AACVGR hajtas_m_Page_035.jpg
db6acc878c8a21eb8191ec834fa9c5e0
13e540dc03f3150bcb93766506863c2f1e8a9ad4
89831 F20101206_AACVHG hajtas_m_Page_058.jpg
acc4f3de219b33efe4cbd2d9c233afcc
09b9c863178c4a2c45d1e16fc7997838325ae96a
91939 F20101206_AACVGS hajtas_m_Page_037.jpg
38f0023308a23dbfdd3f4d0e034282ac
ea4bb71f195d5ea18fa3a2d66e1a525b4e88ae6a
94230 F20101206_AACVHH hajtas_m_Page_059.jpg
60fb760e6185e2c649c641ca54689559
c8300bb43126080d552490e54b84a311bc02a5fe
107731 F20101206_AACVGT hajtas_m_Page_038.jpg
79dd7cd47cc3293160421dcacaf19047
ab0bdf568d5811f1f6dc8054172a7cae3fbc8ccd
84142 F20101206_AACVHI hajtas_m_Page_060.jpg
78e00e7090f1881184f12607fecda4df
fb625228404969ca87b7e37e609438bbc44b50d9
57232 F20101206_AACVGU hajtas_m_Page_039.jpg
764930242f6193d8eea77b8961199a6c
da2ecd73005799498c02c7b87b84c21372cbb282
86227 F20101206_AACVHJ hajtas_m_Page_066.jpg
9ff8e66dbf69bb154e7eb58627943a60
9b9b737ea4f22fb9d31312dbfef0c4660a667514
76471 F20101206_AACVGV hajtas_m_Page_040.jpg
57b538f587903199957112947f2d33c8
7ff70509826c7bdc63347514f7490d01d5a9a098
81703 F20101206_AACVHK hajtas_m_Page_067.jpg
4a72708ac6cd4f21e0a660dcd66311f0
f0113e4a08fd843fa98affdc6ecba6b5034faa61
100507 F20101206_AACVGW hajtas_m_Page_041.jpg
8ab99b4cfbe7a13030641e4cfa0af55f
f1a17457c3453a7de4dc3b34f1f4a0f3bb2d1601
85952 F20101206_AACVHL hajtas_m_Page_070.jpg
362c6dc91169ed97f1b2afe99e1093f2
58ecea6dfb6337c96bfa608d2ff597d26b2d1da1
84432 F20101206_AACVGX hajtas_m_Page_042.jpg
05060fadb6d928b94b6a44cae4df046c
680945641179d584aa4522ea9675c1c3bf973e7b
90077 F20101206_AACVIA hajtas_m_Page_096.jpg
d5ef9f406321ad4d646e64691ca143e5
0690e1faee1e023403424ca720184ebf8eaaaf15
87643 F20101206_AACVHM hajtas_m_Page_072.jpg
3f3bf126b02b9f368b93bce487068ee5
d4eb416d274575cce4fc574ead2b789f1ad9be27
77950 F20101206_AACVGY hajtas_m_Page_043.jpg
bc99922e5ea609ef6152d3001cb61717
1b07c140e1c3c604ca6052812c6356a5bfef621a
93336 F20101206_AACVIB hajtas_m_Page_103.jpg
ed46880cf6190aca9dd409425312728a
621cf9e2f5bce2cf6c549d305f1fab9dc56db903
89123 F20101206_AACVHN hajtas_m_Page_073.jpg
8b37e3fb10d0d8d3723482bf4694bb13
e8f5ac6cfdebe4e9e8ffb17433f679e2c3bfe6da
90746 F20101206_AACVGZ hajtas_m_Page_044.jpg
a9aa637c6d7587a4ed28ea67a85ac149
e1b2e97bf7ce4ada55e257892f3244bbd72b5d64
94396 F20101206_AACVIC hajtas_m_Page_105.jpg
eb86eaeb16571b42fbcace7107029770
acf0a6586d06055b79045217e70c24fa13603109
87636 F20101206_AACVHO hajtas_m_Page_077.jpg
06f72c880eda90e40bba7ea045bba535
e7e9388921a347a6002bf2771342d231144614d6
94991 F20101206_AACVID hajtas_m_Page_106.jpg
33cefa3edad6112a8189b4306869edb2
cf1f2f92f07f5ac6e6a68b6fa94806221b139d6f
93154 F20101206_AACVHP hajtas_m_Page_082.jpg
809353473911589110a5bca2a5227ea7
12f48695da2a036ee16d3f7574e6e9164e29dc80
91695 F20101206_AACVIE hajtas_m_Page_108.jpg
9bc41bfd2f34f7035570bf19c6bdece4
d75398ec68a333cc921ce0e8c73498a2924d50bf
82279 F20101206_AACVHQ hajtas_m_Page_084.jpg
a66cfd59ad84a89f99b25ed8a8732aec
2a6a07e89eee14cae4635bebeb8b472520ca18ff
62766 F20101206_AACVHR hajtas_m_Page_085.jpg
eee42727ed73871137c4207f02d43c3e
f15672c88dd19ad243b2fe0da8acd5186d8a4f89
90046 F20101206_AACVIF hajtas_m_Page_110.jpg
a1cc35e18d516d9da7c58b1762f0ee99
4dff194ddd2509ad1f4974014cea85d93dc9ae2d
25041 F20101206_AACVHS hajtas_m_Page_086.jpg
7612f24b2fff85a9dfc745d649454980
54520aad7883ddc9c84644fe924c1b3c029a0af3
94729 F20101206_AACVIG hajtas_m_Page_111.jpg
d65bf5b0c5cbab62ac7e1736b29e42b1
e30bf82721fee7dfb11a51c7c3957c115ec2d439
46439 F20101206_AACVHT hajtas_m_Page_087.jpg
950545b625935ee965efad8549143084
f92a7e1d0b33927854442a39a381953c6c866100
26829 F20101206_AACVIH hajtas_m_Page_112.jpg
982d32f95ec5cc4b002afe5d3d9cc5e1
d6ef8f5d2df0eae4853425faac8a2623b4cc88e3
31106 F20101206_AACVHU hajtas_m_Page_088.jpg
32de30224d1d1b83b06eb31318803dfc
50074412f55a9e4d32950f704b984ca851cf0e15
57316 F20101206_AACVII hajtas_m_Page_113.jpg
3ff87175c8e30a9efe3744fecaa84e7f
cccef0b78c21f5896fedccca7b983028798db956
19413 F20101206_AACVHV hajtas_m_Page_089.jpg
c8b3bf309703997768740e0cbec81844
a756e1f6108df66914a73b88667af94343af428e
51217 F20101206_AACVIJ hajtas_m_Page_114.jpg
42fd6c61be26f9da9159351c57fd14be
bc52c95bcbb46ffa8209543347d10068dae17d42
38510 F20101206_AACVHW hajtas_m_Page_090.jpg
97f7fe46e7321c2a841f04b8d7f93b30
7a86e8eb61a0eb77b38ed3373c2db96eafc7d9e4
56264 F20101206_AACVIK hajtas_m_Page_116.jpg
3f58a5ff1887d72d3a0790533cbe73aa
dd8bbbdb056c8fd94e5e7d5d292830c2a366399f
50349 F20101206_AACVHX hajtas_m_Page_091.jpg
bec1b14b21b91a440f67da9778e64ee0
6c5987ccbfeb9ebfbcf4b2776512a61bbe8661f1
110323 F20101206_AACVJA hajtas_m_Page_004.jp2
36dbee44410e2537228a62f06dbcaf8a
f7b8e5a99ea70bb5f049f033a473988ad5ce8b56
50517 F20101206_AACVIL hajtas_m_Page_118.jpg
99eb8cf42900a5869e51877aa3b3c1c3
d812fe94e2596de63c482743f7dc71a22c55df6e
93356 F20101206_AACVHY hajtas_m_Page_093.jpg
4b6a7c4cd57bca7aded19de1d2f30e91
f09875d712508bb41bd19e079bc37166f02f5db2
81172 F20101206_AACVIM hajtas_m_Page_122.jpg
9072b28e91690bdf8c81f3d43991a461
3f3c3325eff146d356812bbb71fab0c7ae13afb6
90075 F20101206_AACVHZ hajtas_m_Page_095.jpg
4f64a7f0347352cd736ec8742248e958
94ac043ff103cfb61ac46d8064a0735a7e877fb0
69025 F20101206_AACVJB hajtas_m_Page_005.jp2
2db202175cdbe481b5258bc81be599d8
da5811cef2a11dd1c426070421dfc4baedf0e313
86387 F20101206_AACVIN hajtas_m_Page_123.jpg
bbe70404864868b203491d8f0802491b
df836e2bcad991f8a62271844abf2d2efc987db4
1051947 F20101206_AACVJC hajtas_m_Page_007.jp2
ac0afb27b89410d4f209ee74c53d604e
b0cefa9c84973b2e7c093c2db062ef63ecec17e3
83527 F20101206_AACVIO hajtas_m_Page_124.jpg
37a2a39e32a770494bae8210c7492508
a98ad593b2d8d84f10ca5e89daf33ecfde1ac43c
162489 F20101206_AACVJD hajtas_m_Page_008.jp2
02cca87b53d90b2720e535370379d1fc
8b583fc7e1e243659c27fb774476105836e4c64b
78957 F20101206_AACVIP hajtas_m_Page_125.jpg
efb4f81aa32dcc56ab2cc63f13b66083
1090919ea265ead4d0aa0712073c88a3766271c7
203909 F20101206_AACVJE hajtas_m_Page_009.jp2
be9f9555f3d1af154cd8a4465304a4ea
736febc3d00e6cf3b5032f5887d4023f384eb74d
74121 F20101206_AACVIQ hajtas_m_Page_127.jpg
027310614cf5b2a2257dd86c846b7e48
21ef848748c13a0aa4853640e125979a8b38fa90
1051982 F20101206_AACVJF hajtas_m_Page_010.jp2
93d477fb408fafb805d646c4341b9867
a648cca8fb520597c62defd471445f14ba9416e3
96318 F20101206_AACVIR hajtas_m_Page_129.jpg
bb2d8b7b0998847cfb1a5c2e672476fe
91a57855bcf16690597be9358f97f127a0a5821b
103924 F20101206_AACVJG hajtas_m_Page_011.jp2
e3feffb993b3b23d885b32f06f7dffe0
9b803abadcc5bc03080d027336690daddeda4195
92193 F20101206_AACVIS hajtas_m_Page_130.jpg
5cbdbe6660cef0ba7e49357da7e9022f
1edbc2d9f8f9ed576cf47c6fcfa897965012e581
55638 F20101206_AACVJH hajtas_m_Page_012.jp2
7f92f6ba24504ed29da04763eabdb83d
9407f124132b0a62f1cbd025d00490fffa82c60e
91710 F20101206_AACVIT hajtas_m_Page_131.jpg
8c3f1455a419bfe344c462c0e593879f
31fe39296ae31c91bfc2cdf759c00032db1cc2ff
118477 F20101206_AACVJI hajtas_m_Page_013.jp2
acebed3886ebd105513e10ee93ecda48
6b37de949450189c05ee118d05db2fafbf7752f2
93173 F20101206_AACVIU hajtas_m_Page_132.jpg
fe5d44f0fa382449e95d9824e481a444
71ec47c0056a6ff1b0add61a6ca6ce71a3fbcc36
114058 F20101206_AACVJJ hajtas_m_Page_014.jp2
b0ae1f8d6decdf6fe12e457346c0fd43
f2028c1eaf810bc53a37453ed5d44bee21df93c2
42025 F20101206_AACVIV hajtas_m_Page_133.jpg
ffa99c2e9431d18a7d0aa86731ae0a3d
d3f2f848b1bf894f8fcdd4b7c9b5e005321b2332
124873 F20101206_AACVJK hajtas_m_Page_015.jp2
5deb86a50744f7348b11826c5d09180f
7429fa1b5804751cfe29fc6a3a0860600a6cbb80
100245 F20101206_AACVIW hajtas_m_Page_134.jpg
8e638d526c7fed36e693513110eac375
ba39b806aff1acf97e9f69bf8bd9ced854c38fbb
115656 F20101206_AACVKA hajtas_m_Page_053.jp2
fff5bfd64d4b9971490eb0a674ecea1a
c99f148ce97cfb578eaa8369ac6f0e9d6df96c98
122814 F20101206_AACVJL hajtas_m_Page_016.jp2
98e33322bbf6c37fc944eeb682232f70
88e94f352a89ad362646aea37b923899512c7e47
102685 F20101206_AACVIX hajtas_m_Page_136.jpg
d36c6b006aedf21b4b27e4ee955cc949
9d5dfd232b05f67671b5609d8f23e7b6d9c20396
116052 F20101206_AACVKB hajtas_m_Page_054.jp2
a86f93ba98985c33fccc88b552944eb3
a6e1704bde48ac90dc051be90a5ba32da15acd05
119997 F20101206_AACVJM hajtas_m_Page_019.jp2
c1bfd2758d918c2b5362b80ed5aaee2d
3f0dad5b88e1ddff8ce8ba6f788dc2907d961d1f
67527 F20101206_AACVIY hajtas_m_Page_137.jpg
b4c6b34e802445d48bded09e44d5bc1c
0fc666ba29f6d4c548aefe9988abdbae80b26aa9
124523 F20101206_AACVJN hajtas_m_Page_023.jp2
2e2c85cbaa5352a6f272f09935abc989
1c39cbe6d9026e39a43bee03a61a9e61cffdca90
5267 F20101206_AACVIZ hajtas_m_Page_002.jp2
707ce5a51d72fbd1eff3fa79047cdd44
b9a00e4cca2f4fb42a0c4aa192578a64c173ab24
121497 F20101206_AACVKC hajtas_m_Page_056.jp2
d982690ba69eed1c826bf61a43bf8a56
2299829c1692c21b9c956fb7d82ac22bfda49c84
124776 F20101206_AACVJO hajtas_m_Page_029.jp2
04a4f72ecc6b8e989a929f964211acbe
d5c80d29ba7d2a6b601a66476087eba4ca8d048e
109846 F20101206_AACVKD hajtas_m_Page_057.jp2
b88a96f9d0b130a38d81af8801a08649
12c01bcbed84a7cd3900c2d3a43f56378ceaa3e9
120051 F20101206_AACVJP hajtas_m_Page_030.jp2
08a2d80a853f2ce06dddc7e6c3712ee9
1f2ce025a2bf3dd2b6895cb6c17cd4c94bb3363e
116699 F20101206_AACVKE hajtas_m_Page_058.jp2
c7a3f15f78e5e6a6aab5f08641df40df
98743e014994ae1b9ac8edf1efde01c862c8c706
1051979 F20101206_AACVJQ hajtas_m_Page_031.jp2
3e75bcbb5e891e34e607b13b53665586
52fdd0ad0486ca2b331adc9dd9a7d47890e564b7
1051956 F20101206_AACVKF hajtas_m_Page_060.jp2
c4214a6196b1e7a9d3ddd2ea6de1513a
f7b0c52c0896260c873b0536259b9c7f15a24656
119115 F20101206_AACVJR hajtas_m_Page_033.jp2
eaa12887bca6d5318b7951a3aa754280
9b78a3adea04b88d87230514c8dfbad101a3951b
48676 F20101206_AACVKG hajtas_m_Page_061.jp2
b2730a66b33eb6f780721460cea63b44
e45076531c480f7bbf1dfc154aeac02d6777fd01
1051961 F20101206_AACVJS hajtas_m_Page_034.jp2
7b7749003cd20dcee0b3c7f889c8db1b
fe28d9968d56450585aefdf65af67a2427b24b6b
348889 F20101206_AACVKH hajtas_m_Page_063.jp2
5636e475e9f33a101a282429cdee974f
4c91db56565367fea837b3e03d6eb5919863ba97
115020 F20101206_AACVJT hajtas_m_Page_035.jp2
48cc55dd17db5696936e594c96f8194a
7e0cc2661abfaefb55fb2d88216790fd0d927037
70463 F20101206_AACVKI hajtas_m_Page_064.jp2
255b7a207b24a73c8135594f4b464be2
fb7574c5ab4cb3f6f95be4be486cc4d4054b5515
73543 F20101206_AACVJU hajtas_m_Page_039.jp2
a02a3bfe7d7a105c83dc91c656f184e2
d96daf0824cf5d349b3d26b48eb22231437dcecf
114551 F20101206_AACVKJ hajtas_m_Page_065.jp2
14e7f9f17975cc9b2ddd6595e7e8c1de
d8c2efbfac278a5dc3fe56418235b85f5e455172
129402 F20101206_AACVJV hajtas_m_Page_041.jp2
f243240d7935f9a2f711bd9d90e9c175
225c8be22dfc041e3ca6f94d65fc31cf0fe160f1
112624 F20101206_AACVKK hajtas_m_Page_066.jp2
93cbd60f7445f586c69c7dd4d16d1d47
20c0f4f0070b608d3604a0a0bdef1f0b2b6a290d
1036445 F20101206_AACVJW hajtas_m_Page_043.jp2
aa9a7246bbf7662a531c30ea1760e142
ead664428d453894dde6c8c2553f1b5e09d323d4
106633 F20101206_AACVKL hajtas_m_Page_067.jp2
47f463f85f39bf57370ebfec8cf93189
496c7a7c0e602a1b8fca8857fee2023b99f09767
1051986 F20101206_AACVJX hajtas_m_Page_046.jp2
8086745f691e6dc4d9b66e81cde0c50e
f590bf29d1e30fb4a8bf79af0e863548fd35f25a
350732 F20101206_AACVLA hajtas_m_Page_088.jp2
2a6f88ea078e3223ea541abec97eaf84
ac6186e272b895fd960489815358fea79d209b45
113034 F20101206_AACVKM hajtas_m_Page_068.jp2
6439b8fa1d8cb8f460a5aea5ffcd8f25
f9adb1805fa38c26edab6b32251838e16ba60345
117133 F20101206_AACVJY hajtas_m_Page_050.jp2
3839f6fee5dd1214121166a8c9f803d7
9b127d28db02957def731994dd7c6521b128bac8
418085 F20101206_AACVLB hajtas_m_Page_090.jp2
4e056fa67cf25812d022ea8226068b6d
230fd893815852924e01c893096a39dd43088921
107205 F20101206_AACVKN hajtas_m_Page_069.jp2
647f1ef30dddf3f02523f69328d9deea
b53e10b3006db5bb4e4c8f28b54c9cf69420adb5
119205 F20101206_AACVJZ hajtas_m_Page_051.jp2
01cb32ac8824b56b726177d07afd1d65
8507570e2fb468fafaa8271ae48e090c253d3640
596305 F20101206_AACVLC hajtas_m_Page_091.jp2
2968183f6c4fa12712f60306a62be2c6
8935df1223fdbad7e182a1dbab82e411c0410fc2
112686 F20101206_AACVKO hajtas_m_Page_070.jp2
0a72d5d7434b8c0769abea463d180a33
aec331dd0090aa3bb0a09446d72c55ea73db990f
113470 F20101206_AACVKP hajtas_m_Page_071.jp2
c530bc63acd66637923527eaaddd2a88
e860bb1a953a4cc1a99fab5b0838e7b77d2691bb
108817 F20101206_AACVLD hajtas_m_Page_092.jp2
b947f6a8b7ee1eb8e334fe439ca02046
3b5dcf9b0492bb54d4b5eea22d8b6846459a7561
115671 F20101206_AACVKQ hajtas_m_Page_074.jp2
455dd3bad0173e4dc0abb6dd80889fdd
755123d2e77f02005796c3bd1fecb10cd7ff6828
121663 F20101206_AACVLE hajtas_m_Page_093.jp2
a83425de489a849b2ec023f6f193e3b9
73937fa750f13dbd43249d6d8d7ece5de35b3204
111920 F20101206_AACVKR hajtas_m_Page_075.jp2
3845745fd980eb13e0e570c6567ff2c3
6a4882c783fe6d9eaa432c5da8ab71bad2e722cb
119896 F20101206_AACVLF hajtas_m_Page_096.jp2
82b01d0cdcd3107327f741a2f6b8850b
481bb8908e84031315ca61e35213d61422ce02fd
113957 F20101206_AACVKS hajtas_m_Page_076.jp2
54724d33e42284bab15ad67f7d58f9bc
179afd48d5df0f105a83ab7477b7e39ac7e0977f
122852 F20101206_AACVLG hajtas_m_Page_098.jp2
a6af275d9ad58521bf479155ff18535f
3da3b00e300d20469f5c1131e9ce49627d19ded9
121006 F20101206_AACVKT hajtas_m_Page_078.jp2
72576c5bbe5df91bd5f83883f5db5e15
3c1ad1c75c2c8ad1cdd8275f9da30844cdf04ffd
120701 F20101206_AACVLH hajtas_m_Page_099.jp2
90aab0a84c3cbb7297bcb2883dca39e7
04d90f43639f38563418caf0abdc6c60d3b78ce9
121338 F20101206_AACVKU hajtas_m_Page_079.jp2
86371f9e7377019a0d510e5ac62db161
fba37d4a5226912cf8d152466ff952a50a512e99
122221 F20101206_AACVLI hajtas_m_Page_103.jp2
59db5702a7cd4cc9fea4de6bd95c9312
962c057a6e8ed2096051018433d56df1dbfd4909
114378 F20101206_AACVKV hajtas_m_Page_081.jp2
89d67e3b0a51962fd16e93ea48fbe307
6d14fdc6539b63ee91ca16594ae472a492eb658d
124361 F20101206_AACVLJ hajtas_m_Page_106.jp2
5811049c0f25eb403972b0d542fcf920
6da525ba4ff2cd65419ca566f2299567787e610a
122758 F20101206_AACVKW hajtas_m_Page_082.jp2
f253b720e79689098f8b7ab08834983b
9c289c63c8b2392bff78c773986dcca1952aace7
122561 F20101206_AACVLK hajtas_m_Page_107.jp2
52e03a50c5973922a1adcaf8cf4c3dd2
9cd9ae11ef36569226931d0c830acc5650167c97
117943 F20101206_AACVKX hajtas_m_Page_083.jp2
c7070b96dc685ae55d3bc6e8feed983f
e9bfc3e0a151b17ebb858b140f5a1664b00767c9
120085 F20101206_AACVMA hajtas_m_Page_132.jp2
55ebf7359f32ffd19fd249c97acde6f6
707629fbb177065a6a41719b9307a32022ffcf7d
117935 F20101206_AACVLL hajtas_m_Page_109.jp2
20cfb8a41a26464cd613ba98db08db3c
352e0ca6f27d3cd8f41edef35d15acb1237efcd7
30776 F20101206_AACVKY hajtas_m_Page_086.jp2
f7d9cba361d690340de95c6b500c8db1
cab1eb44d35f9f69e90e5fc301aed318c608116e
53691 F20101206_AACVMB hajtas_m_Page_133.jp2
903f0dce854fde205c04680762e75158
fb1b4c5de9a752f8f0d9a8b60ff2c738826b48bb
34526 F20101206_AACVLM hajtas_m_Page_112.jp2
86dcd584b408e5cfe54e9644f433b582
2d9aa8c4c7eadc632ec79da7fcc17ab356dd36e9
502645 F20101206_AACVKZ hajtas_m_Page_087.jp2
f35dfa1e766b8b9495d4f955e9cf0930
409acdf5cef0b53aa3dab864af4921eaa28cd121
92421 F20101206_AACVMC hajtas_m_Page_137.jp2
cd288c0ce13ccb1282eba976863fcade
07e9496a26b8128258ac63d5883d13d0c185410a
64126 F20101206_AACVLN hajtas_m_Page_113.jp2
b3f26feb62caa79548d79bdc7b679814
d21a1312159fa762c35599392562688736159a69
F20101206_AACVMD hajtas_m_Page_001.tif
6b28ac70cc2260980c5cd9967af61e3d
d96731f96b5851240c36d6f247b939a94e076347
55260 F20101206_AACVLO hajtas_m_Page_114.jp2
bcde5be2101d2f03e9e676858ffc2e9f
0844f1fa87f4476e5260777f3b97a6f8e5c3004a
62199 F20101206_AACVLP hajtas_m_Page_115.jp2
aa1db250872f04117bcc447f4d406083
1d3efb6cb2874a6b15fc919e9d67963de7f21ccb
F20101206_AACVME hajtas_m_Page_002.tif
36a3fd8beb8f74c8bc125600490634c9
1a42829ab3d57720c0afbda4ba3fd56e108252ff
61972 F20101206_AACVLQ hajtas_m_Page_116.jp2
25c34169e911c356f63d452165fbbfc1
e7e12c7119ed7fdd43a2ea6ee7ba84dfc460bbda
F20101206_AACVMF hajtas_m_Page_003.tif
7a55eeb7b81eb5c4ec0a6e2ffe35b4e2
549d5d570a0f91211240601958f852b1c86014d9
60098 F20101206_AACVLR hajtas_m_Page_117.jp2
f4a26adac24721db47e5339dc72a2f4c
b351cdb8d4ac8d82bb479d6c6115e8cb9b4251cb
F20101206_AACVMG hajtas_m_Page_004.tif
0c2df1e1c396bf041ec6c0485e97d069
5155b6356447e62a79de5ce93486d24036849f8d
38542 F20101206_AACVLS hajtas_m_Page_119.jp2
7ade6ba1768a4e0542bc2a0568f92efd
806419ad8492b5e0b995b250e7377ac1903762ab
F20101206_AACVMH hajtas_m_Page_005.tif
1c48b5699b61d2f7c4a57e1b890424e8
357c8ba2ca1653b80577d1c8756f024a1358e724
89318 F20101206_AACVLT hajtas_m_Page_121.jp2
5942ffbbe9692400c2b098bf30185d9f
1719d9bdb770d30b6660a69ae11abc908c0f1beb
F20101206_AACVMI hajtas_m_Page_007.tif
bbba5135eb45e1bf0f5512d187a7e133
8dea39a5696520cca551140fdceafd357cf410d4
108005 F20101206_AACVLU hajtas_m_Page_123.jp2
be4a9c0da6810b6c9c0f0e2950f9f91f
f308056e089b1e44ee62a7786bf54e3d82ad0085
F20101206_AACVMJ hajtas_m_Page_010.tif
4e714bf9cfe79541f507c463c560a0b8
b1bcb21084747d47f7ef71dbc56d578085ed1726
105587 F20101206_AACVLV hajtas_m_Page_124.jp2
9a2295acc8e3747f055c142fd957f518
09d4bfd818159d9f7cd215d558115afd39e9717b
F20101206_AACVMK hajtas_m_Page_011.tif
79c4a84b46d1ad9dfe55566b8de023f0
04084a97216af32c00aac6bd5e05d86d93e0e532
101325 F20101206_AACVLW hajtas_m_Page_126.jp2
86e6b770bdebbd474c18a5409f8a1b77
79c63c91f88b97bfcd9fb86ba59f6576a045c7b6
F20101206_AACVNA hajtas_m_Page_036.tif
da5a6f14bda742cfa1be15961cc148ae
0e7e4f7e0af4aefb6ecde175ee4c3e8dbd23d48e
F20101206_AACVML hajtas_m_Page_013.tif
c69d57056b63837996c19283f7da5cd8
aeab0d233c502ab06c814e4a17b9c89648f91892
97660 F20101206_AACVLX hajtas_m_Page_127.jp2
c531349589b51a5f19c5039ad7111d69
d7c0fb3a58af371a59994028386c1a6adc541dee
F20101206_AACVNB hajtas_m_Page_037.tif
86c6fb90522551b3e3bccc555cb804c4
5ae85de988609ea59c9b4f0b6dc6808c646fcb82
F20101206_AACVMM hajtas_m_Page_014.tif
4edd3fddecd97ceb2e1ba40a69dd2707
cc5228dd7f943c99731b2ecc1eb2cfadd3e2a9ff
105968 F20101206_AACVLY hajtas_m_Page_128.jp2
20a569fbdb51b62ae94d147d7914081f
9471622128de19be65ed47f786e4a9bc2068aab9
F20101206_AACVNC hajtas_m_Page_038.tif
53c62779b8ae581d57c69b7d92118fe7
f84abb8adf95ff01137139c33bca7363147284d9
F20101206_AACVMN hajtas_m_Page_018.tif
486b568e6157fec2d0cdbdcb518789a1
59c16a692779028d3f2e8b30f0518613a148c686
118334 F20101206_AACVLZ hajtas_m_Page_131.jp2
29ebaa8f9c32c23936c711a0e3a3cd6e
280f985e1aaa841947f63fc04fe1b02b0b359c44
F20101206_AACVMO hajtas_m_Page_019.tif
52c2a2512c059b1d6785253ffd02af16
24051db1796fcc6ca5fe6870e0cf27140116ed0a
F20101206_AACVND hajtas_m_Page_041.tif
543e3be3a67d7ac00b7db48c1a5ed39b
2e251e6546ac378fa8d4f48080f11ce43f9e838d
F20101206_AACVMP hajtas_m_Page_020.tif
2ac827fe539c15b2786170824776c539
156d0989222114437cc4d63b6cf57980d3acf30d
F20101206_AACVNE hajtas_m_Page_046.tif
b177baa99f45c73a17e7ba571192817e
4521fa641af0ec77969f30b5f1434f94291b5cbb
F20101206_AACVMQ hajtas_m_Page_022.tif
44b80563303850385d78db6b482006a0
99bf8867c93d6903a9e15535dca62210379d130b
F20101206_AACVMR hajtas_m_Page_023.tif
6035dbb27411d7815a1bda73f5ce81f3
f5731fc9c751c8b5c54fe0bbec7100fab03f6262
F20101206_AACVNF hajtas_m_Page_047.tif
345a542aed91b0d1239461bf0b2a81ac
9a68ffb223208f37576ff6dfe66f0279385f609e
F20101206_AACVMS hajtas_m_Page_024.tif
c8e62a6e8109af97271669ea841ee31d
3799a65818a2d1c74eb62d0b6e30b1e626f7f044
F20101206_AACVNG hajtas_m_Page_049.tif
322f568e4a54bc135d17947b56d3820e
82930e5af48e438f320d1ab634ae9488dca6dd12
F20101206_AACVMT hajtas_m_Page_025.tif
9b6a0a95c08676d2ab0bcfc9184e55c3
d3e466e9a03bcf3193efac8b69810f8a2def64c4
F20101206_AACVNH hajtas_m_Page_050.tif
4d756d523226447d655ce1e0cfe510b1
beed87ccef5879196b7250d71b0caa9c3ab97704
F20101206_AACVMU hajtas_m_Page_026.tif
cffbc936d2a7d8dad88727d397686394
15fd320981e48d43c19125b39a6129f638197056
F20101206_AACVNI hajtas_m_Page_052.tif
50843c637717dd1a7785dbc483b3331b
e565dd7a02c1b36b7aee6a12744cb1e89fb85fc2
F20101206_AACVMV hajtas_m_Page_028.tif
14cfbd30f9f1ca11a607fd9654b270cb
4feebbeb759d0930f0c2b936ec7e4a2646d25089
F20101206_AACVNJ hajtas_m_Page_053.tif
3293132be658d3036f9d5c1624dd672b
2f13d3f5fd0b883167241a5877e0354cd815f4bf
F20101206_AACVMW hajtas_m_Page_029.tif
b8a121afc4d9fab51da89c7d66a64289
91975843453e858b0717622efb36ba26e4c7ce01
F20101206_AACVNK hajtas_m_Page_054.tif
f1837720cbe049e19de07e0421c2de2e
c152f856b043f625d95a0672b48e70b70fb99a95
F20101206_AACVMX hajtas_m_Page_030.tif
567703ea395fab1d1ebed2d444a20d6d
63aa398d5bcd8781ec91736519e2562cfacced23
F20101206_AACVOA hajtas_m_Page_086.tif
1cab822e5302d5f0fbd030036cf9c2d8
1c3f80c7a4e1e641ac2e5521789c5d6f606f1305
F20101206_AACVNL hajtas_m_Page_057.tif
d48c62110d09aa80601538ea40529961
4347e2412a01aff60d4cacc12cf2ea8c25e200dd
F20101206_AACVMY hajtas_m_Page_031.tif
5e944bf6080a9d61057013a4ee26187b
e4ba9030292df90d94b0030374098508d12dbf1a
F20101206_AACVOB hajtas_m_Page_087.tif
8d44d1046d5a92eb88ccb06466785811
f798c16e60208918ff98688f488a0ed1983e351f
F20101206_AACVNM hajtas_m_Page_058.tif
92d1a0786c7d60f5aa648993484b3a65
5aec0f981f6ce519c9bdb82bfe45404fa8443941
F20101206_AACVMZ hajtas_m_Page_033.tif
6a99e55c0f5fad08fdc9fc86a10b2ac2
8abc1d6c311aeac96275d507bfb780ddfb6f5515
F20101206_AACVOC hajtas_m_Page_093.tif
ac93c24554cb3f2263f30c2cb7dc0fd7
9861ebe4e95d14afad99ed0a6bbcb26a01ef383c
F20101206_AACVNN hajtas_m_Page_059.tif
c02ec60fe12bdb65617b7e8da02af870
98b73cc586a9e9c124385b9e6deb3982c982d101
F20101206_AACVOD hajtas_m_Page_095.tif
01e5acc79927efca3beb5df2b3c308c9
5f6b1fb5f09c8b78ba5e799447c23ec651b3aade
F20101206_AACVNO hajtas_m_Page_061.tif
2af902360f3854a1c15f6186e5c1cb7e
1c30b003c95dda7dd2503bfd30ce96697c964d96
F20101206_AACVOE hajtas_m_Page_097.tif
e7f8da79804c849b6f7b6a461dbc44b7
a3dc04f72be2041bde9bef31e08616682ee71ac8
F20101206_AACVNP hajtas_m_Page_062.tif
1afebf7609931e40f8d2b2ce009decf9
497212d5a5f4fda59c6133bcbbe2977356a1576c
F20101206_AACVOF hajtas_m_Page_101.tif
e8ca86fd68de7e9e50f9576ef2fcaa87
5799fbbac981e0404746898c32a323c9c23bf5f7
1054428 F20101206_AACVNQ hajtas_m_Page_064.tif
ae7bc406d2e46d995c715ba9ff19dd0b
f1b44aa00a61c71a68b08f52ae4283a81494df2b
F20101206_AACVNR hajtas_m_Page_068.tif
10889c4293b02e9d93a493770c549659
8b28f97fa8710c151c04f60a546e8bfc468bf552
F20101206_AACVOG hajtas_m_Page_102.tif
af822328313dd29a650cb00fcb867a3a
8b1291577b5d03342a55a68021b510d8ca661ff4
F20101206_AACVNS hajtas_m_Page_071.tif
be0c23d95d45c676ae57ea5e10920812
1d80c1a1d5169c7c9ca7a9bf59bb10fe8f07c1ef
F20101206_AACVOH hajtas_m_Page_103.tif
66cf8f1d096a4ab5f1de80127e7fde06
5c095e48ddd61ea13f76e853931fcab1b753425d
F20101206_AACVNT hajtas_m_Page_072.tif
e4cba70fabab9ba97f435cc5c3696f40
7b9cc740142bfb7b89751ca07898dec05c9d6637
F20101206_AACVOI hajtas_m_Page_104.tif
156fd16cf8a64f87d046681bde9ed8bd
9bcc17c0d42a276da7c2c5b8948bb42dba01f7c6
F20101206_AACVNU hajtas_m_Page_073.tif
2d5a13aaba8316dcb3e2e35891e2e25f
87174dcf2f19ef63c22cbadbe0aee290fdcb3d82
F20101206_AACVOJ hajtas_m_Page_106.tif
e2fe9ff3d6c58d9ff79893dfbc631038
7a8bc163d8725d2b4c88efad033dab8f957cd52a
F20101206_AACVNV hajtas_m_Page_075.tif
1c020432406dc23eb29a3f15d5be6792
14824826b5293b710188b7e2e933dcca5feb36f9
F20101206_AACVOK hajtas_m_Page_107.tif
eecdeb417dcb3ff8c367e53ab643d19d
8de19247893a7ab0ef653b8b43c4883d7963c9eb
F20101206_AACVNW hajtas_m_Page_077.tif
9e8bd07e613f35e5a0612a440041dcd5
8e51e3358a5a7e3499d434e2abcaae6ab2f8e1e2
F20101206_AACVPA hajtas_m_Page_134.tif
b19f07a7026a524d2c4f358980177f45
18b9b94f0fa7b991bc90969ebe23863ea1955ab8
F20101206_AACVOL hajtas_m_Page_108.tif
9f6b6e4b37c36ca2dcc9f1ee0e0dd033
260e0d8c0dca3fd1f5a22af501c2c114dd268f0f
F20101206_AACVNX hajtas_m_Page_078.tif
7dba02f46ef71d72aa98e62a9b5a31e0
6c16ad490a4eefe34f98dfdf82426bda7bb88d36
F20101206_AACVPB hajtas_m_Page_135.tif
0c7b6d58ad18b7d6d5c6f08ff8d9c1d2
a8ac36b96cd607a17e4e54f81114f1670c74a043
F20101206_AACVOM hajtas_m_Page_109.tif
fc8fa9334a12f0f3db9c98d01425de3f
a29bd3433c6ed25138873bdbf036ba5551fdb614
F20101206_AACVNY hajtas_m_Page_079.tif
2e954b2fdc002dd63bb55b68de978f03
4aaf552bbde575bcdc8b1c8a01e1e7319792d228
F20101206_AACVPC hajtas_m_Page_137.tif
96006ee043e7b417535867f4de6ee018
135105303abc7062e09ce901cb47cf3077a4f417
F20101206_AACVON hajtas_m_Page_110.tif
daadbf8fb04f1861d6bd801aecd5d1fb
9fc1fdd380c1bae4be841a8dfb484272fb52841a
F20101206_AACVNZ hajtas_m_Page_083.tif
afb202b7180718e3294557664a63028c
d5286b85f486fc365ac0e73a6c325fcb8b053ecf
882 F20101206_AACVPD hajtas_m_Page_002.pro
d76c2b58fd3f244e7910186cfaa39ddf
50fc518a8dc78e2340501fa824e3c95b61edab86
F20101206_AACVOO hajtas_m_Page_111.tif
d997c24f524ba3e2bf273ce0b82b227f
8099eb2c182646963e0c9d8133fa065f3f1e076b
1506 F20101206_AACVPE hajtas_m_Page_003.pro
693901631b6f3a501b22828cfc6d92e1
602cdbb882c976ee6bc8606d20835bcfa7e13d39
F20101206_AACVOP hajtas_m_Page_112.tif
8687be30783724289afd594c933897d4
20c03caf31227b08cc85c18cccc12e0e7baa2252
30978 F20101206_AACVPF hajtas_m_Page_005.pro
4fee5daf5184c575bd93a530a6f44248
2dc145b45f1bfc692e8d0562843969637a760db3
F20101206_AACVOQ hajtas_m_Page_113.tif
244fcfd90ae645ae984cd52a6eed3148
bbe18d8da45be424a1a16e9f5e3acd38cc544ffe
47230 F20101206_AACVPG hajtas_m_Page_011.pro
0772f3a2eb67e3eb566183b11ad135c8
af9a3b9de8475ca12e22c8f327ec192054d5db1e
F20101206_AACVOR hajtas_m_Page_114.tif
94970fa1e97dae1e63defa71645f8dc9
de4f93f8067fd31e3772e20317886054e27af84f
F20101206_AACVOS hajtas_m_Page_116.tif
aa60155620d6873477f156412c6e84bf
ddd8c54bb232d7aeec590ed1aea53a5f4ad2bc64
51997 F20101206_AACVPH hajtas_m_Page_014.pro
940ecebdd8f6496e5e65b47438dfaa48
7a5043cd169c9aea992197311128380db59e24f3
F20101206_AACVOT hajtas_m_Page_122.tif
239bacdd498bb91de1c07babf0f2d006
833a0776779ab39d9ac4074e797333380fdda0a4
56919 F20101206_AACVPI hajtas_m_Page_016.pro
75f33fb98a41e1f2e45f2f6200865f81
070acc31cc37d24128acb9fae5cfe5341ada2947
F20101206_AACVOU hajtas_m_Page_123.tif
a791e58018281b166b13119c19b49c94
8ceb537477cdbafbc1666b443fff376dca17775b
54367 F20101206_AACVPJ hajtas_m_Page_020.pro
337bfdfa277390e704974a009a3b153c
320b5ef7da03af33575495479e0b8e80c3864073
F20101206_AACVOV hajtas_m_Page_124.tif
58489a6128ba2754376761f73ad92ed1
8e8c13a86eea179ff0f428ad16bd2c585079e6c3
57430 F20101206_AACVPK hajtas_m_Page_021.pro
784db0febd9d6de50ca33f37421f5dda
8ff91f20b1270a238ef8f3078266f5ac4907ab6a
F20101206_AACVOW hajtas_m_Page_128.tif
e009ee44668291a618f3b65dc22f23da
d55c43bf73d70fa09ab04fe576ed62f3ae41b0ce
55248 F20101206_AACVPL hajtas_m_Page_024.pro
40d7ec81a49bd95d615282480e78095e
18c4879dd2441cb7a1405ced6c7513644c83e7f2
F20101206_AACVOX hajtas_m_Page_129.tif
c5ec143d6168ed8da5a21236a94d62d2
616dc840f2a77c85781234bcda64bcae8e5ea946
53182 F20101206_AACVQA hajtas_m_Page_048.pro
6b6081657f0896dce4ff094e884d872f
21391c3d362b5015249f8de17d442d1dbd6099bd
56917 F20101206_AACVPM hajtas_m_Page_027.pro
415f01a8db965d701f1b3720221e4418
7abc49fe9a7e0201bf6ed099e4e2111a39a43726
F20101206_AACVOY hajtas_m_Page_130.tif
9c7e3b63cfdecf0dfa142e20d2968368
0418d6221269b84ddbbd7cfa6b9784764795fee7
50212 F20101206_AACVQB hajtas_m_Page_049.pro
44f6eff52b41a62f903391c1792a588c
8895bf78468366801586615a0989dfd8f22ab59c
53119 F20101206_AACVPN hajtas_m_Page_028.pro
af26df954dffc5f402f31f93545f23ca
716dd3bba0fc222369c9da0cc7878be9c7bc2909
F20101206_AACVOZ hajtas_m_Page_133.tif
7656def8bc289e5d466c655378490c03
58ed0eb756b2d5cbb804cb2e7b516ac3a730007e
56420 F20101206_AACVQC hajtas_m_Page_050.pro
372b0ab0e9c395c4aba444367ec165a1
1db0c1c0030e8aff23d370cf8697e465928af39c
56976 F20101206_AACVPO hajtas_m_Page_029.pro
b644db42fc7f8299c845a37ba2ceced8
a9e7441768f5246e689d69fa0e5a27762b9d62a0
53935 F20101206_AACVQD hajtas_m_Page_053.pro
7d0d38722d477a2033d19e6669f1b636
492d86112c403e297fd93b917809dd878aa29665
57226 F20101206_AACVPP hajtas_m_Page_031.pro
e2bc25d079bd692445db821f0606bba1
bbb903fac0de17bafcbb54efce990365e272421f
54145 F20101206_AACVQE hajtas_m_Page_054.pro
2040b846b1fd15ff55de48696ea2663e
3eb12eef9bb1cb224213718b226d5a724dbaad0d
53438 F20101206_AACVPQ hajtas_m_Page_036.pro
8590880697fc6e1597833f39efb30b97
251b3d2b802a7475e03193af4d4f7e1f6fcd26fa
66384 F20101206_AACVQF hajtas_m_Page_055.pro
f72091f7dc1508f44da9a01fa46e7a77
7b9fb999d5b7e1b83d2c469cfa958a93dfd76db1
55924 F20101206_AACVPR hajtas_m_Page_037.pro
25ec7664d17349b862eb8ad1860d2543
818c965b7fdc52693a63e83ba6772a4ac8264e4b
55563 F20101206_AACVQG hajtas_m_Page_056.pro
a86cf1a0c7c87be84581083b0eab8867
2f37a258bbe38a1f747ea1d7b3b230dc1e2b0dab
65560 F20101206_AACVPS hajtas_m_Page_038.pro
dfbf1cf5e1383db9f057251d07fe6bf5
5a0b8c27343fc548d25be4b2fe6c3ff593ff916b
51493 F20101206_AACVQH hajtas_m_Page_057.pro
bd286f625f3db889dbe9ada7684ac800
3fc05320ffa7339cae88864b92f718d0e30d8968
32442 F20101206_AACVPT hajtas_m_Page_039.pro
48bd4ab54505a6b03de92433f4fb10e0
a02dc461c90a70bd39a182494192d20efd5a34cf
45600 F20101206_AACVPU hajtas_m_Page_040.pro
3503ab5d1cff7fad8b11d1729308050a
b7ee3b68becb5782ca911c876599108569c1b09b
54062 F20101206_AACVQI hajtas_m_Page_058.pro
2b6339f61a19edf7e444fcadfbed4d2a
ea0d4da646a57cbb04bcf3a09b5e361319ead2e7
61650 F20101206_AACVPV hajtas_m_Page_041.pro
7a914eada5a435089a3a4d9af3f19a05
1f1d2e5e5ee66c7dc4bfb104b86852c6fa0fa4e4
22701 F20101206_AACVQJ hajtas_m_Page_062.pro
f0cba27285d61d6f482e3771e62595d8
7873880f13c2216ac88bb541a883810a4909e2f3
50314 F20101206_AACVPW hajtas_m_Page_042.pro
75cf6da38b186776a2c6c1282d8d03df
3ce5c6cbeb2c47fb1643b522079bdb6212efaa20
53365 F20101206_AACVQK hajtas_m_Page_065.pro
b31f6665bf5755989982c53cd4eef621
345c9d3717cf99f08d9b94196bffebfdef0eb699
50188 F20101206_AACVPX hajtas_m_Page_045.pro
e2a96b2fdb613e8b1919c43237e8df2f
5a2cd5e054671991e8982cd94586302551d69225
11499 F20101206_AACVRA hajtas_m_Page_090.pro
b0b9693299f05b39972eac47bef5840d
0b0d1918f5b875ca3598e10b23bf0b5ebb4cc0bf
49049 F20101206_AACVQL hajtas_m_Page_067.pro
949b4d99998d0aeddcb9f344dd2a4c9f
9ff393bba8fde2c3013dd82f8b689d9b89bdd26b
53882 F20101206_AACVPY hajtas_m_Page_046.pro
470e4b8f286692a42326f6967c81e4f9
9f12b3e551cee1b97bfd0546b0e3bd0036625fe9
23350 F20101206_AACVRB hajtas_m_Page_091.pro
2fa8ccab8ac636ea390dca08f5a98e51
e233a2408fe14c9ae30a3596fc8c1b4cea603b20
49198 F20101206_AACVQM hajtas_m_Page_069.pro
f1c31fdda99ebb532143cfd376e6d8cb
9b409c2ceb784dc85b582075365b240acc18d9f9
51892 F20101206_AACVPZ hajtas_m_Page_047.pro
c3b5073276cdc4c328d773d3c41dfc87
824cdc3555bb632767524e49d1650cdaf0692e88
50756 F20101206_AACVRC hajtas_m_Page_092.pro
034d9fa6cf217739a74461d11e811b4c
62a653d64d8016345ed6536b714b9f399159dfb1
54109 F20101206_AACVQN hajtas_m_Page_074.pro
315ed977f3a89f152153d700036d7c64
848881aaeef511e8b39e2c1545878a6e322aae17
55406 F20101206_AACVRD hajtas_m_Page_096.pro
ba07a983103d2c1d6fd7f5e30e3467dd
a6c0b6c0afa77ed480238e53621d286a4b61f897
51315 F20101206_AACVQO hajtas_m_Page_075.pro
630e69dc3c3c49a9ee32880db1626303
1f97f66a79fe927221e9b3a5fd9bb73cbb0cd9f8
57363 F20101206_AACVRE hajtas_m_Page_098.pro
517ce268c09ae50467ad690473de552f
b68452b9948f2b946faec5e98d95832864f40f1f
51646 F20101206_AACVQP hajtas_m_Page_076.pro
46d90793f947902cef5748f987d71be3
26969b0b41681266bb7422ced64d825a8ebe75b7
56557 F20101206_AACVRF hajtas_m_Page_099.pro
c6b1f792d3abb0f0a4923e21c5d96a59
ee896fcea3e626a7e3402aff44e976e4af2828ea
52763 F20101206_AACVQQ hajtas_m_Page_077.pro
88335497544cf66b653454250ffcb014
9c2af30a418a8f36ea6a03212d8995b939e9785a
53734 F20101206_AACVRG hajtas_m_Page_100.pro
0226e50c60e0a2ee1f7fbc22bf47a077
3b52eedca3a4c6ed9a39eb097b928db614fa7dad
54997 F20101206_AACVQR hajtas_m_Page_079.pro
5c366f4ef80456f587c8a2db0c0bf7a1
32357bee114c49a1a934593b51fe66eb8209bb5a
52588 F20101206_AACVRH hajtas_m_Page_102.pro
97477555715599c07bf002d2588cc854
52ecc33bc752762fb0f45904f2445cf019b86bb8
55304 F20101206_AACVQS hajtas_m_Page_080.pro
4c23a869ec800e58998ba3d77cf976ec
5495a1d2b18a6bff01049e6b899d738bd621d05d
57461 F20101206_AACVRI hajtas_m_Page_103.pro
96c6f2b1b5c2e9db2502f3b709fb8316
9cfcb6a245aa544d8dda2b7dbb0869374c53ccb0
52053 F20101206_AACVQT hajtas_m_Page_081.pro
80e3a18dbb68a4c5e028c386fca7c670
8c95cba0ae52acd7063e0153b846fface753ab3b
56906 F20101206_AACVQU hajtas_m_Page_082.pro
73e0058b8d8c09c3c97f4b6a6edb8d2c
e8b2845130dd6fe804375641e9525edefae9beea
57478 F20101206_AACVRJ hajtas_m_Page_105.pro
799e6009a6a88a420ba9ecb5f751ce69
66e21a508032671d7176d4edc197f5bfe6e7422e
54071 F20101206_AACVQV hajtas_m_Page_083.pro
b4a1e94b5511063b0de337e2fc034c1d
c3700d28b5ecb4bc3182e06d2bd7d5d483d7ddea
58114 F20101206_AACVRK hajtas_m_Page_106.pro
b44f654bb96153cbf11d27d3d4024709
b4ec4f602a5fd16c3358318f55f9d650651c5af2
50433 F20101206_AACVQW hajtas_m_Page_084.pro
f5d09969c1d6fe82e44eeb304e7787c1
2879227065b291daf97d794be3069059cdc4f28e
57267 F20101206_AACVRL hajtas_m_Page_107.pro
f98c5e87cf7ce0bd8c217a16ef8c0b22
e8a659bfecfb1a7962f2c0a755de96f8072c9817
13601 F20101206_AACVQX hajtas_m_Page_086.pro
6571c58b773a02ede9db6891064df19b
03c70678fe9d82378bcc1abf86779b1ad4f72754
43498 F20101206_AACVSA hajtas_m_Page_138.pro
a1a07316338c94bb9b5e904f4269b544
733f5b9b49a29900548cdb9279eb2337265867f2
56483 F20101206_AACVRM hajtas_m_Page_108.pro
dcdd063b2e143ca64175e653a5532727
f98301c545b47d748bb7aa0afb907450b73fd9c9
16620 F20101206_AACVQY hajtas_m_Page_087.pro
ff06e5f26ff05d648fc11e94a1eb16f8
f8e633d6dae20a99830c075cfc5fabcf458801a5
486 F20101206_AACVSB hajtas_m_Page_001.txt
30a5b0486ec01a7f6cd57031d7f3e105
912101af201c71a17515de0622c7329b4a8458f0
27106 F20101206_AACVRN hajtas_m_Page_113.pro
c31a2206a8b86c8bce32060e4157e4d2
df0da1c8d9da30428a69d159e8e2901203c9a42e
13528 F20101206_AACVQZ hajtas_m_Page_088.pro
cf3c2cddff934e333ec52dff842afa57
65316e99f727e974b425335119a917ff62095115
90 F20101206_AACVSC hajtas_m_Page_002.txt
c60526d143bd0fd0486bc9c20e66ab05
1a37a9f09109efa9dab1d9d193e853ea4126a687
21439 F20101206_AACVRO hajtas_m_Page_114.pro
87be629155adf4aee65cf5f89d76abed
9021bb2f5e733805b51a1215026188b692b70ad0
104 F20101206_AACVSD hajtas_m_Page_003.txt
6ea8ce51b490807de27efe0c0d709531
364a82608d42b4f6f9d75d440e1ffe444fadeeb0
36997 F20101206_AACVRP hajtas_m_Page_116.pro
54dd727675c4b02d0ab9278d0abe81b5
acefb24aa67c72020d00e2dfcd30dd72faea65bc
3535 F20101206_AACVSE hajtas_m_Page_006.txt
f227dfbd4fd8eb708437c7e40df928c1
f45d66eb6f3075243fc53d400b5d653556392fd7
38620 F20101206_AACVRQ hajtas_m_Page_117.pro
f08380917d61ede83bdb61e35e73fff9
af35b495d369d591005c62fe4b002d934506eb2c
3804 F20101206_AACVSF hajtas_m_Page_007.txt
4f5dc0f145de16796ac7f4c4a6f8e37b
df14afb805344544b5f34e3a06555458f65d787a
34229 F20101206_AACVRR hajtas_m_Page_118.pro
cfe0c5c0ccc8faaa4c310884d005b1e4
9de35def03470086233dd792f50e43394136fc24
211 F20101206_AACVSG hajtas_m_Page_009.txt
5cd1531805bb0ed745c2d7c5ee5d2b18
55ca30af5d2aee3390833eacae7224a836675f57
10317 F20101206_AACVRS hajtas_m_Page_119.pro
db9557fc25c7ef63ee2fe71caf422f28
2818426393f5c8b2b8887b8da88d75270975b1da
2078 F20101206_AACVSH hajtas_m_Page_011.txt
59abf569efa943dd9bd5b4cddaad6069
8b789cfc0fc8edacc0995d3518a94925d094be3a
39983 F20101206_AACVRT hajtas_m_Page_120.pro
dfad1419a7b35ddee418413d27619b2f
9a2e0867353206da622ddf9127bdb81141080e17
966 F20101206_AACVSI hajtas_m_Page_012.txt
3f2e634d8360311e49fc6921e20418c8
2485d3c3f3b3bdc83f27254fc4788d22d7ecd143
50568 F20101206_AACVRU hajtas_m_Page_122.pro
78e21758121888477eb74668d32cc9e4
7b1fae0daa49ec668efa39c551eb602ab4d8350f
2209 F20101206_AACVSJ hajtas_m_Page_013.txt
4f926bc4e407e606324b0e29bcb7b50c
5f16307e24efeaa19144e4c5941b6bb1f9baee5d
52481 F20101206_AACVRV hajtas_m_Page_123.pro
25277293e556951a8f795832a6d78c2b
66bb6c5ca97e1394f98d1789f9636400f3fe9ef5
49750 F20101206_AACVRW hajtas_m_Page_128.pro
c3818946d266d55bd28e276126ad2301
61e6e18ad89b127effac0d2f7f14d8ae009f5904
2271 F20101206_AACVSK hajtas_m_Page_017.txt
28566a425e764c532d341e634536728d
3b773d61b7cddcead8152c3a597670b6a4288478
60572 F20101206_AACVRX hajtas_m_Page_129.pro
80ef5a18e6f26c837b58ea0d047acf32
c4f6b111758a2a912e5aedf1824ca67a29a01121
1989 F20101206_AACVTA hajtas_m_Page_042.txt
4b280c5e02bc9ed53513409eef3b9c9b
21b3be918317323e841099fc246bf1377ffb7f72
2100 F20101206_AACVSL hajtas_m_Page_018.txt
a1e137b513fbe07db195c82df5963f39
647bb66d74ed14bdd35bbedfadfcabe5894d3db9
56728 F20101206_AACVRY hajtas_m_Page_130.pro
b39476c4c24977d42d351c9651f488c0
79cf9b8471ee3b2edb4bb08ee8aa2951771ec7b2
2154 F20101206_AACVTB hajtas_m_Page_044.txt
e47863d8ba06e308c06b7e8132b85326
ee03e625325bba5f591ce1162a03797505a0c40a
2171 F20101206_AACVSM hajtas_m_Page_019.txt
56831549dd81e6ad15c1ac8227ab2148
0f4cf0b67e49febf371fbffc1f569a3fd311f6a2
62557 F20101206_AACVRZ hajtas_m_Page_134.pro
512f6f00e3ccf3d1949592c3e706eda0
b030e4e2f58aab8726eebd23217ed0d53ccbc914
5043 F20101206_AACUQA hajtas_m_Page_128thm.jpg
fd14bf3e9a12615b4a32bec616090f86
0b56ae9a4f10b9004dc7b30b755820deac0aa56a
2002 F20101206_AACVTC hajtas_m_Page_045.txt
7fd11ed4c57cecbc15dabf998ceff907
6336e3bb021eb9d22acb7ac50346b5d885379c79
2135 F20101206_AACVSN hajtas_m_Page_020.txt
4f94fddb6ae5d9e04aa2b7c6c3dd390b
5c34f35708f925f79d8854242f1b0b30b1d61e22
27797 F20101206_AACUQB hajtas_m_Page_095.QC.jpg
d392e72e1921adf30084b95abc07bfe5
929c32257175026b864192a69966e0c4f0fe0666
2138 F20101206_AACVTD hajtas_m_Page_046.txt
d2f885b524886b114647ba0883443325
b48b5ec0c745434b041bacbc406dba2123f39a19
2246 F20101206_AACVSO hajtas_m_Page_021.txt
0143202f4cecb21beefabc9cc1debcf5
68974b0712d2a57c57b2d91b65569fa64952a150
89999 F20101206_AACUQC hajtas_m_Page_083.jpg
e5929ef165613c284c7072f968288b97
08f3a251ac81a72a1f3fe2a44b59aa7c0d779c2f
2059 F20101206_AACVTE hajtas_m_Page_047.txt
dee1f01bc153bd02301db11ba8152b7b
8b4bdacbe9483729a7ef0256a7e41d5f0b2da92c
F20101206_AACVSP hajtas_m_Page_022.txt
276333e2242dc6b9c0858d8b89c58ff4
50969c19ba4415131d0c6dc16d5792607aa8929d
89373 F20101206_AACUQD hajtas_m_Page_074.jpg
ee35dc1e54dbb304747369abb64d536a
7453f64e93fe7bc0cf5b69fa9ed7f26f930e6395
2104 F20101206_AACVTF hajtas_m_Page_048.txt
9f264ac4e0481c8bba6f3f4886271aa9
fff5946ed4af839f7bba38c7f595a9ca48075a83
2198 F20101206_AACVSQ hajtas_m_Page_024.txt
32a3c34a8f6951361cbc1a65bcc5e8fc
9b569d40d417328c8e1f28a6c6ef2af3d5ba9fb5
F20101206_AACUQE hajtas_m_Page_016.tif
34bf37f1887271df36b1e67805ef25ab
172dcc13475f543abb46237c3de93fdd30d664d4
2218 F20101206_AACVTG hajtas_m_Page_050.txt
902ed19f6437343eb54bf7388019f263
e2f3209231933bbeb5a82e3732182faee68f0168
2227 F20101206_AACVSR hajtas_m_Page_025.txt
6cd34e2f43a291a571f7b731bfa5141f
0ab22d896086edb43d50842cfb51fd24219efb71
58871 F20101206_AACUQF hajtas_m_Page_034.pro
2c42546e5fa89dfed8737a8b22b3e491
e5c04f090228fd17cc0621394029f249a5d28d10
2206 F20101206_AACVTH hajtas_m_Page_052.txt
57349ee4a2796ad0a690fff0c3a413de
2f50acbc716bcfc423deb2e112ce2a343932d207
2267 F20101206_AACVSS hajtas_m_Page_026.txt
19692773a8e21be1760ec624e8453088
d5f56c04e14a5b56bb53c3ceb2ce8ee5f5e3aa87
123177 F20101206_AACUQG hajtas_m_Page_021.jp2
448b391ef53b89485d081e414f80a016
4a349a596524955fed3abbca13ea658c846c3fa2
2704 F20101206_AACVTI hajtas_m_Page_055.txt
f59666e1dae7da1edd9d1ac33c1fbbb9
ca7d637668d2e7ff53fcde89455ba45c84ffd7e1
2229 F20101206_AACVST hajtas_m_Page_027.txt
3b20c2c780b5d2c3c81aad75b8dbda23
474f2129ee3397a8d55b886de68ca35fb58e351b
134806 F20101206_AACUQH hajtas_m_Page_134.jp2
4934bec67ad370755fbad6a4a3de3330
c3d88678dad0092720d35de2df6692281f5befa9
2030 F20101206_AACVTJ hajtas_m_Page_057.txt
0306649bfecf76dbe7eb713447ddfe02
a40bf82ce1257e3f5fa3e0a217cb10373176537b
F20101206_AACVSU hajtas_m_Page_030.txt
dd0bcab4efea17b4fbbc66e368633b85
0684f7f2fa9a011b7a2087a951a330d5ba0fa8a6
109021 F20101206_AACUQI hajtas_m_Page_042.jp2
45615418e73c1ec15b7bedc6cd83edea
1ad293e1acf09c258e16181894b6ed74eb3da9e2
2126 F20101206_AACVTK hajtas_m_Page_058.txt
b534efc3fc43ed2f87d9708e258b1307
0b6189fedf632d036398a947d3998682f3b53c52
2247 F20101206_AACVSV hajtas_m_Page_031.txt
d3b87d865a7f5170e45cf72af9a7e6ed
162045f5a1da67a367ca9e7accd31388b3fe48e5
2174 F20101206_AACVSW hajtas_m_Page_032.txt
0aa3990795ee473e3a00d234252208ee
682fbda00c3d840797ab38e98e0dcce125274495
2230 F20101206_AACVUA hajtas_m_Page_082.txt
ed9e9aa4838f4d970247615109835dd1
3f161d732936046ead301601afd9f4d3a6bdc321
97241 F20101206_AACUQJ hajtas_m_Page_034.jpg
1b631df98ea69a22b1583354d84c1056
6a8f235924d91c6bb4f8cddb53246e7bed0b5fcd
2260 F20101206_AACVTL hajtas_m_Page_059.txt
7407caf29c095381308f9fa5eecc7a3d
eaeaa9f823cc40ccd98cd8c8a65fa1ec9884f404
2063 F20101206_AACVSX hajtas_m_Page_035.txt
e83dfc63347b6e635238f0dc31780e9e
d509898b35fe9b0df4830e2e179b979e41cc40e0
2130 F20101206_AACVUB hajtas_m_Page_083.txt
2c43bb86bb04c7fe2560f5505446e2dc
bd2058499a7904e2988c29744aee274693648d31
1802 F20101206_AACUQK hajtas_m_Page_127.txt
4d9e8a56c342faac19776545444321f3
6a34e1d4b03f5edc5331532914c8ffc76cdad9ff
2014 F20101206_AACVTM hajtas_m_Page_060.txt
6f26235ae9917bf314a322db1aa4ab0c
562b3048567d2bc35864da01265479454775ef03
F20101206_AACVSY hajtas_m_Page_037.txt
31af1d1c106fb716759206cc49c8d8f3
13da6611741bcdd2f69dbb8509c70f86fb1da89b
55071 F20101206_AACURA hajtas_m_Page_109.pro
51f0842f62fb3f088312fe8663e15ce3
f7888831b5ccdf5eb3e8bd85cd262b7389ee5c80
2004 F20101206_AACVUC hajtas_m_Page_084.txt
e4328a451439820c133dce3311b7dcb2
95199bbe2054d2ab0b12d1890a675a97b8195da4
57761 F20101206_AACUQL hajtas_m_Page_026.pro
eee597b3a74defa5488f039881c572be
d6d76a91f47388a8e48bd99d136813443c64eddf
1093 F20101206_AACVTN hajtas_m_Page_062.txt
6c1f54a3d70eaba601964a4f8b2467cc
193c0916895a629ad39b5ee64ba330d66fb064ef
2303 F20101206_AACUPX hajtas_m_Page_088thm.jpg
afcd3d43f14e7db04bcefa60af05fca8
a9505a8bc7d199335fe53cfdc90c61827d21ed5b
1956 F20101206_AACVSZ hajtas_m_Page_040.txt
50b9fb658cc5132199c1822713a5d1c5
40f0467c9c3310be06487f880e092999b1c9ac50
594 F20101206_AACVUD hajtas_m_Page_086.txt
de5eecdbe1c1ec10875b22f49c612063
e713bcb8e25d648211e7885aa98989dcfa0dfb03
93797 F20101206_AACURB hajtas_m_Page_025.jpg
708f2606e521f41469a48551e8b60b48
1f313e3ce6f48f06d87f642c13159bd9685febb8
91476 F20101206_AACUQM hajtas_m_Page_056.jpg
7f034f7d3276f104c704365f2a5cd0de
e4b5d76ed28b12b0b96ea6ff72827a0b70e1e42c
803 F20101206_AACVTO hajtas_m_Page_063.txt
5593b7a6636d99cd9675995ee6d097b6
db68312de183020a39434bae51caac661be70998
6392 F20101206_AACUPY hajtas_m_Page_102thm.jpg
4fcf8a3704c721ef5458679eecec1158
287098bee4f15d5e01e47f56c5b79279525b77e3
714 F20101206_AACVUE hajtas_m_Page_088.txt
10e2aaa83e45c26459dbdad203653f77
5bb7c45b53a5eabc1e12cd06b483854892a6e2ac
28528 F20101206_AACURC hajtas_m_Page_014.QC.jpg
6f19e66ee2c445150cd52f1cc92e5296
a68f616fb3d4d803f229af02a6f4a0f70b9b5e6f
28659 F20101206_AACUQN hajtas_m_Page_134.QC.jpg
2d566aa72def751c26068793c10c0fee
fa1e6712357930b7748b35e44a03a4df35cb2630
2216 F20101206_AACVTP hajtas_m_Page_065.txt
8c245cdb7042043e4d6f5dca8ef971dd
503c1fdd1f944a58daad465da1afb2602af5d774
1248 F20101206_AACUPZ hajtas_m_Page_005.txt
3d0f14341ca96d90e61184e5e3710bcc
65dcc846f3a10d4ce890f229098900d73e3ab529
558 F20101206_AACVUF hajtas_m_Page_090.txt
5a492b2b4ed29fa38b154cf3c204b6f2
25239938601a67d8bb88948d23a3cdba8d85430f
41549 F20101206_AACURD hajtas_m_Page_137.pro
8f995459edea0f6feeac17d9b5a9c7d1
f88ad0582b1bb7f3cc2b55c2825669959139d796
27143 F20101206_AACUQO hajtas_m_Page_065.QC.jpg
9203854fd8c85f54ae72859e51a681c6
e1a228159afcdc712e93c06fbd04ba224a6791a8
1950 F20101206_AACVTQ hajtas_m_Page_067.txt
4e10bc1f805edf9edc38d065ef62af77
f3aa8b396600e17766e389dc47007c35c4ee51ab
27006 F20101206_AACWAA hajtas_m_Page_048.QC.jpg
00932355536239b2343ba922ae58033a
8397e788222d42d846ecbaefbe563cead994cc8a
2256 F20101206_AACVUG hajtas_m_Page_093.txt
027c72b5b807df24cd0a0a04012120f3
f96f494f6524b82522545cce61c6a0129b3fff42
2298 F20101206_AACURE hajtas_m_Page_034.txt
650c559af0d6890f4625775020a6b5c1
fb2916a7825ac16cff414f3eb45bfb57df38b8c9
82671 F20101206_AACUQP hajtas_m_Page_092.jpg
113dc892b208a88d73946c47d26b9adf
7fef2809c48115b34b721dfeb200d82afbc1d305
2067 F20101206_AACVTR hajtas_m_Page_068.txt
022b77e940747e53012e7128c6ae515c
6eead8f24266e30c64db22af7f60c5339ab32e2e
6710 F20101206_AACWAB hajtas_m_Page_050thm.jpg
d29eb3573ff24b57ac8e6925faf3ded9
f03cf65a4f77def80647d538c149e7c1094884ed
F20101206_AACVUH hajtas_m_Page_096.txt
f55c36dc4f13ac79b55438c6d9d423d8
e80543ac2bc16949258b5d92c99740275d730045
92023 F20101206_AACURF hajtas_m_Page_079.jpg
b188952e6458e51af6e31b38c6d3c04e
9db02406c5dd0af645e6d932a7be076a604b998e
126901 F20101206_AACUQQ hajtas_m_Page_129.jp2
5b5dfc2280f51b8440c219ab0d10400c
2e3289c57befbb4970dffa1ed2227cba50ca73f3
1957 F20101206_AACVTS hajtas_m_Page_069.txt
d2643c1741800bb6e08cc2bf977c2263
9359c825c169c87efb8a8e0a08b5c9d1100eb163
28367 F20101206_AACWAC hajtas_m_Page_050.QC.jpg
f11c99c85a119dfe8e1e3d808118797e
f88393914ab1d6fce4657f8ef666113fab6cf959
F20101206_AACVUI hajtas_m_Page_097.txt
c73b36bebfb076d808d0b07a97d4a49c
c7e0bee57dc4068b95cfe96cf87b0c33b3228ca1
6772 F20101206_AACURG hajtas_m_Page_019thm.jpg
2c103d274a3bb38003f450475cde57cb
238a5827e72c407eb19982d65eab937fc33b7559
19051 F20101206_AACUQR hajtas_m_Page_113.QC.jpg
693580d29c3c8edb5ab3c4dd913ae802
f7ae2710be1cc1881ffefb207ecb3e2da9d72acd
2042 F20101206_AACVTT hajtas_m_Page_070.txt
67bccff5638e8d59bb2a39aaa05a3b9d
e2477968e3569fca1c27996f4fff14a19f957c34
6527 F20101206_AACWAD hajtas_m_Page_054thm.jpg
40f40445b83741d92928ebf13252a1d9
96963211d2e05ff7cff5947d43588ddc3ad5e20c
2273 F20101206_AACVUJ hajtas_m_Page_098.txt
fa290fa63a24d183d604862584e52f35
3b35ffa377ab5f5088a932d3ca63e09bc04ffaa7
1701 F20101206_AACURH hajtas_m_Page_137.txt
747392bcc4080076c5ed30f2a4a7558a
3029d11d261caa7c92ae74cdf3ce27eabee08130
27849 F20101206_AACUQS hajtas_m_Page_070.QC.jpg
83fd37c5d1dc448ca16739aef07f6eb1
84f4b73324ee25d6123aea27348d13a458e1c6be
2084 F20101206_AACVTU hajtas_m_Page_072.txt
5b500f5b235f9d4e60c4ee43a114bb43
78ac6142c3c5814eb2a927e13539c15444b44622
27301 F20101206_AACWAE hajtas_m_Page_054.QC.jpg
85e8b6abd436fc5c88cb43c42589a43f
d86898bdb25ed74fc44684d8b45e9d90ddc9ad61
2193 F20101206_AACVUK hajtas_m_Page_101.txt
745904f06de3fff675997ff87a1a97a4
e5382bc190a34195318c245f131b29a5d16ba128
3959 F20101206_AACURI hajtas_m_Page_062thm.jpg
3b2260d1c2f743ecab1a1137ccfaef4b
ffdd10ec9c3d4a8f12d7a647030842e7418126dd
29983 F20101206_AACUQT hajtas_m_Page_029.QC.jpg
a13b93d0e8e0b06af0590301d44e9d2c
feace8cc7c801d1f2fd84632f77fe750c81b9611
F20101206_AACVTV hajtas_m_Page_075.txt
401c55c35528d694520dff6786d10637
2bf9ee5f54d5237e958067ce07d061161478d8a2
6235 F20101206_AACWAF hajtas_m_Page_057thm.jpg
bdab05e732bd2cc57ae44c84a6927a19
9700624eabe24e234c80fb56b3f659fd9779ee65
2068 F20101206_AACVUL hajtas_m_Page_102.txt
8e5dfdb8fd173fc7d12b7c2c9ab1f4bb
6cd61f6be2fc5c2645c3d037881e2e9f68b88588
F20101206_AACURJ hajtas_m_Page_056.tif
e38649f2c58c40cd445b65612986a315
11f0f3806021e2f82df14cf1c3125b9ac5c52179
94687 F20101206_AACUQU hajtas_m_Page_023.jpg
b0c809a96eaced89463092164ef1df66
795732aa4a59eca01a77b079a4f0ecb9bb6666f3
2083 F20101206_AACVTW hajtas_m_Page_077.txt
df0b8ff50b29a806efd26d36edd1e0ed
db682499ec50002b2e9af8e0609c829226664488
7043 F20101206_AACWAG hajtas_m_Page_059thm.jpg
b73063833628379e25c29bea36b117f7
9b94f8e49d281855186b82b744a4031c68b8fe0e
1893 F20101206_AACVVA hajtas_m_Page_126.txt
a7be29ab574838684a99476e5eba37a1
990c57c6c9844199a06990517cb4acb02d1e98be
27976 F20101206_AACUQV hajtas_m_Page_041.QC.jpg
fcc27dc5930fa674cb31eadb21513549
7831acb80af6e6c6f697fd56c7844035941a91cf
2157 F20101206_AACVTX hajtas_m_Page_078.txt
e680177f754c2e0eab1a469ff00b3b06
d0dd6398f297213ff9b4ad2b51f161b5896495b0
6203 F20101206_AACWAH hajtas_m_Page_060thm.jpg
b87d59bed793257302fe1af104e71c6e
51bf00c85232e4a175a069e6d94d3d0e862422c5
1954 F20101206_AACVVB hajtas_m_Page_128.txt
90a8e4bfb87418ecce6331283203f35e
2bb46ca3884bd6637e69cfbaf09ca522ba692219
2282 F20101206_AACVUM hajtas_m_Page_104.txt
726219f0f9e0b2851172222fabffa2f2
d751a6c54fce103bfcd5e82359cc86ccd242d4cd
3001 F20101206_AACURK hajtas_m_Page_008.QC.jpg
fa093952e03451d23546adc345d05ba5
5087b319cd42a4f02bc820acb051f625d7cf3984
57013 F20101206_AACUQW hajtas_m_Page_051.pro
9e9beccd1aecd6756d0ffe5b53e860e3
06859684678ccfa0da10f605005ff003cf0bc4a5
2162 F20101206_AACVTY hajtas_m_Page_079.txt
4473e676905af1552782e0ab891e352d
a5c6d8c734c3e889a556b497938251eef3233000
25815 F20101206_AACWAI hajtas_m_Page_060.QC.jpg
9f8328338af4724b4a75abcbe86a5049
ce6e7746844392b2c5218f58ac187190f77bc595
2197 F20101206_AACVVC hajtas_m_Page_130.txt
a5a49826c746b8758c9fcd112d15c27b
381e33ff1299625875ae7bc924e0fe91c7b366fc
2272 F20101206_AACVUN hajtas_m_Page_105.txt
38c460ee90740daaa15ec9dc020a3e0f
526f6ad2b1b758b24c907732ae4853cf4a6aa87b
56294 F20101206_AACURL hajtas_m_Page_022.pro
f4f2bdeddb64821054bd59f47a5e6e34
d75778c0f925e29c870165c9d33f30948234776c
F20101206_AACUQX hajtas_m_Page_090.tif
5b87ccbe17cfdf624fecefc4c60fffb6
8edce0b7801201b0c7c5e107945ef9be57a8ca9e
F20101206_AACVTZ hajtas_m_Page_081.txt
5b396951c3f8d4a474393b1a68952860
d56c83e282215aa26697d2c0d7a1a7c6e235cd02
2109 F20101206_AACUSA hajtas_m_Page_100.txt
e370f5768da7b644128a5406989da1a8
c40c31bb4f4fae02f7303dad367a025c549582a6
2895 F20101206_AACWAJ hajtas_m_Page_061thm.jpg
5286243527f73a31e3355a880b9045fd
22a5cd694c7d054d8067b5665083808d47852c8a
2161 F20101206_AACVVD hajtas_m_Page_131.txt
3a51bd71332c0cd113570afa622fc54e
98955cfbfca45897fe1222d68394c590accccdc7
F20101206_AACVUO hajtas_m_Page_107.txt
83457177c58058cc0bf3070df8e13ed5
564d0be316a788f9af337d39d331effcfe75b74a
2514 F20101206_AACURM hajtas_m_Page_041.txt
999b8d6c7b6362e82aac17e2efaf7cf3
5c02939f0b729c128e1eb0308acedd7e7a03fb9e
99122 F20101206_AACUQY hajtas_m_Page_125.jp2
f0462600d42b6a3b0308b0e8bd1b407a
269b39f29710730c530b229c734277ee4365ee6c
F20101206_AACUSB hajtas_m_Page_118.tif
590e9c7d7fd334a2dee0cfccda7f6d16
2092aab425d3f330b7a75b52b6f030b97933c2e4
8777 F20101206_AACWAK hajtas_m_Page_063.QC.jpg
5782cd6b9e190a8d523c240016c2524b
5dc5ddf18877bfe4666e99d26aaee60df40a0e03
2201 F20101206_AACVVE hajtas_m_Page_132.txt
5503dee2724fa9edc6dd88970f03d0ac
0888ce95a01c1bbec5bcc7ee8a10d2d42512ca20
2212 F20101206_AACVUP hajtas_m_Page_108.txt
931f096a4a88fc5742cf25c98986ff40
0c68f7e079239e471dd5a22809d46dbe30028207
17383 F20101206_AACUQZ hajtas_m_Page_114.QC.jpg
1266d02594805debc4674dce185992b2
55e7faddc08a4de8ab4c296e825c2c20277ecd41
2183 F20101206_AACUSC hajtas_m_Page_056.txt
d16ba39babb9604a0b93945a6a0fbdb1
13eb1449674ec2f25369aa2deeddecd8108985de
F20101206_AACURN hajtas_m_Page_009.tif
d1254d02bcbdfb2128a5ae1a462eafa3
b0e126cdc67baec3a71f721caf00520ac27d5dbc
3286 F20101206_AACWAL hajtas_m_Page_064thm.jpg
944053d0bd5454185b2c608189514662
323ed82af357a59e1af22c9573d1749862aad511
2526 F20101206_AACVVF hajtas_m_Page_134.txt
793a393af4bd1df7f8f1c8c32935e86d
f4f060906d543c232aeec5a1649ba6476f07c17c
2173 F20101206_AACVUQ hajtas_m_Page_110.txt
15f9c8457fa7cbddb677743fa9c2baf4
bb75f3adc8e419547743c603bb84fb13bd83662e
66273 F20101206_AACUSD hajtas_m_Page_120.jpg
e8967fa44f0b09b2eeb9e7e190bc8cf8
3013ec551da9f40b4879b369033f79bbb22efc05
2248 F20101206_AACURO hajtas_m_Page_103.txt
0e440f558eaaf7541f2faa4527dcfeb0
df20b4df228ab9e53c2607a9f8beb757b35627a9
28288 F20101206_AACWBA hajtas_m_Page_108.QC.jpg
f33317629ea097881d2c853bb4125c67
e9c0934241e44fddb921d7ac2e282d313d70c641
10364 F20101206_AACWAM hajtas_m_Page_064.QC.jpg
d5fb2b907ad6f1e6b5ad9dfe75ca93d3
253b8cf1d60a005152731aa62180735025da0682
2512 F20101206_AACVVG hajtas_m_Page_135.txt
f09ae126a6f6be7d1ae7b3a8b1b94766
6f8895fb08060870d0af3b931548f34816d5cd06
2326 F20101206_AACVUR hajtas_m_Page_111.txt
ec84dd4c8bb12be5abdaf5e2ab454aec
628c8081c90bb72b027a6921480e5bb53af903ce
F20101206_AACUSE hajtas_m_Page_065.tif
54becff52a329d6036659c63863d00cb
a6f4b80d31d3f9197cc1ac0031c51b35ad72a919
59645 F20101206_AACURP hajtas_m_Page_119.jpg
019c0d0fb54d2fd22c4d5d70dd02dc5b
3b49040bfa23bee4fb1af118fc1e39330837fff0
8603 F20101206_AACWBB hajtas_m_Page_112.QC.jpg
6c8b268afc846920a8904be4a024d680
8aa690cb011759aef6d59e42c8b8fe12c7ad8fc1
6298 F20101206_AACWAN hajtas_m_Page_068thm.jpg
a1b92367a126d1281a73fa963550d06d
8f2fcee7d4a78e49702c814cd6abd5b7ea6ce2d2
1778 F20101206_AACVVH hajtas_m_Page_138.txt
6ac44adacef120a91260a866ef78b64a
9a27745f6b013e6e88304f08b4115a2ef24696d3
579 F20101206_AACVUS hajtas_m_Page_112.txt
1d4af587886ee214ffa5e77042df7ab1
59087eb8fe0f665bd1f8d46af739969bab44f3cd
6126 F20101206_AACUSF hajtas_m_Page_070thm.jpg
c6a465259fd532450a7a86790ffd706d
d3991444fda7c238219f0d5f00b61bf071a398c1
2249 F20101206_AACURQ hajtas_m_Page_029.txt
a48a1b6dcc96b84d2e71840d566559ba
f52e8a4e3d7fc3bef56b6348f8792ab6fc111d25
4656 F20101206_AACWBC hajtas_m_Page_115thm.jpg
acc87cb103fc0cfee0d63040e1414d55
06681d0c0573052997360f8d29db291e1de44a90
6034 F20101206_AACWAO hajtas_m_Page_076thm.jpg
7a29f8fd37928f8bc8f6542421cceb86
929a55f750c151cae44f282aa0a9b9356a1c705c
432012 F20101206_AACVVI hajtas_m.pdf
a7bf2f26d515ddff86046c84e3bea2e2
609cd07409efd5202865f2179586898be5838735
1194 F20101206_AACVUT hajtas_m_Page_113.txt
0c54f9b55df5989f32bfe9cbf880e124
a7799c495c07b395b6f7639c0747de01058cae4f
2435 F20101206_AACUSG hajtas_m_Page_063thm.jpg
f0d99c35ffc8bb263c92b59318db0e77
c1c3d78c37ed384dcfb37bcaaba94f28edad9adf
64690 F20101206_AACURR hajtas_m_Page_136.pro
37b8e81e3cfe041c5ce9f500f8231873
7a132aab5842ad52961c4aec9fbcfccd718d766c
17787 F20101206_AACWBD hajtas_m_Page_118.QC.jpg
d70ae61598f682894bea6532e3c8709e
97a3681393001da4271fcb22c6433e2690b11679
6517 F20101206_AACWAP hajtas_m_Page_079thm.jpg
8bc1c216095d5809b2539557ac3ed839
c6bbed6af08fd0fe7fcda0f51e3f0711365fbc35
4256 F20101206_AACVVJ hajtas_m_Page_121thm.jpg
b72e1de47b662e9330512daecb090706
08d3b77fcb4c3f32de4800a15f1fb5ee794a04d1
974 F20101206_AACVUU hajtas_m_Page_114.txt
10b08ee49a15a1b15bef2ea3217f57fa
5203075c966f62c7f60b0583c59592f7f2d34031
20705 F20101206_AACUSH hajtas_m_Page_001.jpg
73b154dc9bc6f4ab70561c8c4c45f7ed
8e6d60512548f16c391b1da917cf2f9252ab1e6f
899 F20101206_AACURS hajtas_m_Page_087.txt
7942a427851e8d77bab5f90c79e9f1f8
0832fa03c838eba06ab15a159584f80792d6b3e3
28482 F20101206_AACWBE hajtas_m_Page_135.QC.jpg
f1f6b921b607101375e882586785a95d
e233facc651c00d7fea6fdd7ee637e62d7f677f0
F20101206_AACWAQ hajtas_m_Page_081thm.jpg
b829bbb066d63ae24fb55b85fb463228
8627dbc9c1aff55ccf9e739c8ac29666a9ff926c
1079 F20101206_AACVVK hajtas_m_Page_002.QC.jpg
1acc816bc92e44eab82743299fdcbb19
ecfe6a144409258bce9416500d4384e53e99a758
1383 F20101206_AACVUV hajtas_m_Page_118.txt
ef2c3095531c3b81fd1c9cbf1bdd596f
bc7be13009bdf667e26031bd7f59c6848f4d7eba
36668 F20101206_AACUSI hajtas_m_Page_085.pro
add24b4465878488ff3f9f3995bea821
0c1a9707357d08a4b15dde238dbc189fa29e2c5b
88185 F20101206_AACURT hajtas_m_Page_046.jpg
c364b84c241b39c3808ffe7ad42a37ef
b2245d7ba5582920dab5ff323c95a5f15afae06f
19760 F20101206_AACWBF hajtas_m_Page_137.QC.jpg
0e561aeb8780a4c23841b4e6efe466f0
9a4dfb25e3975bbbfbbac203417a0c8184b6a97c
25788 F20101206_AACWAR hajtas_m_Page_084.QC.jpg
034030983262fbffd465eedd72322f13
6ee544e36a4c20b0ec40cf73f945d784bfc67054
6785 F20101206_AACVVL hajtas_m_Page_080thm.jpg
d58145abdcc189b391d386a9bf298e24
587d71042473b0f424861713e3b82122bf9e72a8
1635 F20101206_AACVUW hajtas_m_Page_120.txt
b252c0b9742106f5016861f7f1f018f5
9ea1a11604354fa055f66d99e69c87f009a2c581
23933 F20101206_AACUSJ hajtas_m_Page_130.QC.jpg
81cf0388e6f883967c2a79593f2cbd4e
48288eb95e7c9149fa9eb0391e5ac42aa4eabf66
2700 F20101206_AACURU hajtas_m_Page_133thm.jpg
408d0e72ad9bc8592adefc63e2362af3
cf7d194097512b590da3006e8f97b6247e845e1a
20129 F20101206_AACWAS hajtas_m_Page_085.QC.jpg
7c30ef4a42df2093b42e5ffc9beed285
0445add1296e3d933128ae80314f958217055621
4725 F20101206_AACVWA hajtas_m_Page_085thm.jpg
238cabb81ce608d33bb81cedf8c29a16
a04f7fbf5ffcfe6c3d4fade743649365734b57c3
3777 F20101206_AACVVM hajtas_m_Page_090thm.jpg
90ae376552f73ecdb399da4e9f825cf3
89cbcc2c996b3196c1743d18037bd93d7ffd4d7e
1671 F20101206_AACVUX hajtas_m_Page_121.txt
a55627220a21594246cfbc07ba577be7
83c3a7e0fcb479cf4323c27d74969b91d6abc6af
7119 F20101206_AACUSK hajtas_m_Page_034thm.jpg
4021d18e2c952e5b53c1a6d3eed71c59
bcf22819386a60e489cdb6324276360fdf77352d
5807 F20101206_AACURV hajtas_m_Page_008.pro
d74fb2142d643cca65d55a58fe1fb80f
e96d102469bc79fe0478625b70fb499560e0e20d
8902 F20101206_AACWAT hajtas_m_Page_088.QC.jpg
f7cbec7f7930b7ae2317de79ba90cc88
22196548d792e9cb04d9a2ba0e476d168061ace2
31017 F20101206_AACVWB hajtas_m_Page_034.QC.jpg
01449576eb186a5fae563fe4f9f4b42c
ca0e7f20f3ac26a4e112d9bc8c5deb18be8ae35e
1959 F20101206_AACVUY hajtas_m_Page_122.txt
e5553909c3d2fc6055eec38e94977432
c7ceb00738b22b6dd4d5c2e1bff8a0a4addd8636
26790 F20101206_AACURW hajtas_m_Page_049.QC.jpg
cc64e605e8d6a7be82bcef8fd317c216
5357961d02a1cc799e69952cb89de25f2ea1a95b
12696 F20101206_AACWAU hajtas_m_Page_090.QC.jpg
fc77d425ae65b3bc1466b08a9d42eba1
1ccb0646ee07a241884f7721fbd1ab8be8b9c7b5
524 F20101206_AACVWC hajtas_m_Page_002thm.jpg
2168442866ce82460fa58342c17dd91f
7bf2d642047e5f051ec9564bb8e4f279285cf369
6721 F20101206_AACVVN hajtas_m_Page_096thm.jpg
dd6e7eddfece2e99f61228a97161e45c
34eb3336e8d3accfb955da81ddbcce709a089bb7
1995 F20101206_AACVUZ hajtas_m_Page_124.txt
8e6e6a4c8c860243d147e21d2ea3436d
008ce2e09467fe712f5efa4d30907b865da1d09f
F20101206_AACUTA hajtas_m_Page_027.tif
fef4bb2271a5bde1ede1560b7f14d941
49b802eef6207a51c8a1cec5eaab32aed3a86bdc
90106 F20101206_AACUSL hajtas_m_Page_094.jpg
6abaf0b52e584ac2affc2745103728cf
c87c23708e72a99a9cf2843d87d29e32c69f318d
F20101206_AACURX hajtas_m_Page_051.tif
6fdee88579240a538058ce53e93c69b5
0e0e915f5b93cbe5d0fcf8198f38929f8cc6f43d
14801 F20101206_AACWAV hajtas_m_Page_091.QC.jpg
c0a5aa3f5e2cc1d72a4411b3fc63ee70
6d9dd7a751856acdb7720de3d012b4b098ad8285
29703 F20101206_AACVWD hajtas_m_Page_016.QC.jpg
3463d0d980438be776ab24e86adb3dd9
966e6029d24a0a5efecab10b9ab51029779a623e
6515 F20101206_AACVVO hajtas_m_Page_042thm.jpg
35a9cf6d3b145a6f9f811ed0ea18e827
7d194e50866aa5dbfa1df9cbb71bf66681b38456
2233 F20101206_AACUTB hajtas_m_Page_016.txt
7094e4b1196106964df9404c645d1b0b
240638c0504351a20ac4e66f765cd5dc1cea7c97
122015 F20101206_AACUSM hajtas_m_Page_104.jp2
9cb0f845578c9feda34174afae0d37bd
7caa157fe978e084ccadd9470cccaf8ba6a3d30a
116843 F20101206_AACURY hajtas_m_Page_073.jp2
45d87facab88bb50c37108584512d0c8
126b9e3e730d44be104faacad11ccfccc3c8433e
27735 F20101206_AACWAW hajtas_m_Page_100.QC.jpg
068f22600d573f5b2a47a3c53f554aec
18bf227eb50bfac96ab8963e38c73969017f35eb
6700 F20101206_AACVWE hajtas_m_Page_051thm.jpg
56dbe68a81ae26a767e565fabe246cb2
6c0efa471e394cf6fd94b97001f673d747f8e37d
28933 F20101206_AACVVP hajtas_m_Page_058.QC.jpg
92a1337c1745043b42c603dd9fda69aa
3ed2d1c1dbff8cbecf754a6f644b32bdb8566638
47423 F20101206_AACUTC hajtas_m_Page_126.pro
4100d9a152c8a579cf6477bbb36209ac
a68f86b2484f516f073b6f0489ad9f18fa56260d
7074 F20101206_AACUSN hajtas_m_Page_033thm.jpg
21fdebb99e7aecf218ff88b7ff255a72
8f243379c5862679b1c58276b923336593f73f41
91696 F20101206_AACURZ hajtas_m_Page_101.jpg
ca03833d15bf7e03299601a97eddbaf4
9c097133c592131c915f9028b54adcb445518995
6866 F20101206_AACVWF hajtas_m_Page_055thm.jpg
70c7d2bda4c2d5f327ababa43c6955e4
6ad771a43d736cc93044a73194b91ce4c29de569
23558 F20101206_AACVVQ hajtas_m_Page_040.QC.jpg
eed505cb6166183384b801edb5becb51
ae78ab2fb2f96906082ffa8ee765affbcf63de46
6592 F20101206_AACUTD hajtas_m_Page_093thm.jpg
53458963894f2d9fe04a57849c4d0026
b1a996154d401b94992582827f0284d1afecb0c5
2050 F20101206_AACUSO hajtas_m_Page_123.txt
f741e3dc00fbcba53c35adf6d002c8e4
05cb845ceb5668f1e248ae09f9be110bd714a128
26874 F20101206_AACWAX hajtas_m_Page_102.QC.jpg
7badae4a1d7081f88888a2a287b3c661
7a5f9a8b2ad9733e7a068cfb894cbbf092da6af1
6813 F20101206_AACVWG hajtas_m_Page_107thm.jpg
90854ea1ff41176ab00c404d7ba864cf
c33ed65774c9558f826a4d8b742e2198d4fd71b5
5700 F20101206_AACVVR hajtas_m_Page_040thm.jpg
1139544a37c69ebdc662a2350bda47ac
ab1b4606ceaf85de62d94365a1a5becfacbddb83
4823 F20101206_AACUTE hajtas_m_Page_009.pro
fe36c8c7f5b5668743610d7d2afd4a88
49cb36b97a05d47b34bfac3a7a5016940e80b848
92239 F20101206_AACUSP hajtas_m_Page_099.jpg
d3b4ec3b0e9d54f1b4aced8530f99c60
e52a4fec7b738b2911e16a74d67ebc67fc38c91b
30063 F20101206_AACWAY hajtas_m_Page_105.QC.jpg
0b775fc036b703e311a12ba781db4197
b4891d710ec81f3920a62a713eb8f24484bb9e92
20589 F20101206_AACVWH hajtas_m_Page_122.QC.jpg
3d292e3f8d01b818743a8100fe3c6970
3f3dcbe472bbb936c827a97686ddd03710bd8c4d
6190 F20101206_AACVVS hajtas_m_Page_069thm.jpg
e1ad016090755238c877c407941c3b14
c1e9e221995b17df3f7eb2c4dad9a4f4cc93becb
2315 F20101206_AACUTF hajtas_m_Page_015.txt
732def8492790a83119296a8b1e99ef9
a1b0226e0f3cb2a9a787f38333074bcb39f1e33b
114286 F20101206_AACUSQ hajtas_m_Page_102.jp2
bf354967b752706f6777cdce4539fbff
5da71fd5e9a2ce6d6d8aa0203600805e80c38bfd
29746 F20101206_AACWAZ hajtas_m_Page_106.QC.jpg
62d45ddc49dcc12942c0c61dae52268c
a03a9970af0bbea1dfb3c9f39dc0ef33fba37bc3
5637 F20101206_AACVWI hajtas_m_Page_132thm.jpg
f7d72e152846b3824775c222db85eb9c
b2c213799753efc0022d5a9633fee651ec255ac8
28979 F20101206_AACVVT hajtas_m_Page_104.QC.jpg
6fa2cbca9d8ca9cf919de73e572f624e
6380508933544b9a4fe5da840418fa66a264faf9
123157 F20101206_AACUTG hajtas_m_Page_025.jp2
f3b8fdbe3bb6efde52ffeea4370c3858
2d08d4847199970a258c3af185a9736a1b2c9652
34905 F20101206_AACUSR hajtas_m_Page_064.pro
136a27f8552f0712cb7f529bb53d0018
0cb2e46793583183cd280175bb27df5661647b97
6277 F20101206_AACVWJ hajtas_m_Page_092thm.jpg
745a5f016563590cf61dacded255b6aa
d18ccd5cb811dc2eb949fedfedb615faf1649dc5
6497 F20101206_AACVVU hajtas_m_Page_028thm.jpg
6c06d2ca8f09f692742ceb22267757fe
768dd10b8ba1010efa5ee62eaca2decb9e836529
57244 F20101206_AACUTH hajtas_m_Page_023.pro
26d546abc7da86ce9685260ae4865ad9
5049900acbaf5449ec67ba23e576157a2addf8e6
4015 F20101206_AACUSS hajtas_m_Page_120thm.jpg
0f9cb9dd54f6340a6b9633676f468ae3
150a78e950fb4410e4078e0a5f8729778e5628c6
30907 F20101206_AACVWK hajtas_m_Page_017.QC.jpg
c7c51fcff98b922cafae771126e8c683
52363932358eef9aa62946f20d30f161584923cb
28697 F20101206_AACVVV hajtas_m_Page_079.QC.jpg
c9d62ed54e80217ffea490d262d1a885
6efe18dca15f90c9d5e73244f067586a0bf3a954
2133 F20101206_AACUTI hajtas_m_Page_054.txt
a1df4ba08e7caa3507bdeee7af78d6e8
77c2afb3cd45a300e3acc58c8c792893716c345c
F20101206_AACUST hajtas_m_Page_100.tif
769f1ee0648ef9c35c18e0bc2989a40c
8e0af84a91b344db76c443afe43303e60023f316
28409 F20101206_AACVWL hajtas_m_Page_083.QC.jpg
1fe9aaed2be62189113c7fe9311b9c17
cb29c5b1344e82afd09a319111409dbe710e176f
F20101206_AACVVW hajtas_m_Page_109thm.jpg
e53b1068f0b77316c7ce1585da29262c
a0ae24051bd2e7c0174a070f11b49641bcdaafb2
110945 F20101206_AACUTJ hajtas_m_Page_047.jp2
38965779fb5031516ae1e5a7d378161c
c98a0551354385e41c9c44da49b322572a9fd3d5
F20101206_AACUSU hajtas_m_Page_096.tif
55170a61676d296b332e86bc3f56d424
138b9dcbac2a0ac7e4d6a02c7d51414c21fa9045
F20101206_AACVXA hajtas_m_Page_066thm.jpg
9aebc33fa2a46bbe74e42b5574b4db48
39f5721bf19f9d0e16494ec9ae9475864ec8592f
25555 F20101206_AACVWM hajtas_m_Page_043.QC.jpg
e3031c6a1fa7b1b2950f10d209068dac
05a0c6eb7f04915c86625a5abfc6c0b7541537f8
6425 F20101206_AACVVX hajtas_m_Page_047thm.jpg
fbd95bfbe5f368ea5e52ab0a2805e213
27c4e96e039803d46d9820e6f6c63789d804eabb
F20101206_AACUTK hajtas_m_Page_048.tif
670dd5ce87f3c0a0f766158dcffdc2d6
88371c3631a3407b898646af3919f117dad39d1d
55243 F20101206_AACUSV hajtas_m_Page_118.jp2
347a830ae3f1e51c8245e35c6ffe38fc
fda90df8017313f1758e9f60d00425ac7833e7f4
28343 F20101206_AACVXB hajtas_m_Page_052.QC.jpg
bdf365055c216e1ca1130eb8cd9c2881
92eb485e843dfb9b35bcbb8021472f53153ff8a0
7058 F20101206_AACVWN hajtas_m_Page_136thm.jpg
741622245660ef0fafaa8e8d70e58ecf
ba13170b2e2527732e40d913ec211f90f07a78cc
28885 F20101206_AACVVY hajtas_m_Page_021.QC.jpg
a4dba1de8afc8bab61ec455a96390a66
638fef929f6806fac51ede88d444b71b50a82c72
2191 F20101206_AACUTL hajtas_m_Page_033.txt
30da3fe7d22898297d3bb64f7c7d7025
f43b37e531ffa92784a069dd12e83637614a93ee
93419 F20101206_AACUSW hajtas_m_Page_104.jpg
cdf3a6ef0762ab934d7746a9c76987a0
454da7942f2535c65b1d2df9cd5d64b14cbf72b0
1415 F20101206_AACVXC hajtas_m_Page_003.QC.jpg
bc434dcd40e024c0a4a0d583e699dd20
42dda3ef6494143d8b94c4d9b2d7b0ec1186a505
11053 F20101206_AACVVZ hajtas_m_Page_133.QC.jpg
a343d92ac773491b4c3d2a21a0baa1d2
1246ee00800b4fd8eb05de1e530a57e322389acb
2155 F20101206_AACUSX hajtas_m_Page_109.txt
688b18d9cdf0a241f464cf60e9a807a9
2ed157f04c0aff0eec64ceb8329deb361d2fac3a
8034 F20101206_AACUUA hajtas_m_Page_001.pro
aca5a645ac87b44a8d63a8370fdfdc20
f61978274fbce2de9f4d9a17553685f903f85612
6537 F20101206_AACVXD hajtas_m_Page_071thm.jpg
e763ee2368f43bfc9b486585802e935b
1a1ec4ed5edc75a91ebbc037a3a4061e96ea0e2a
30118 F20101206_AACVWO hajtas_m_Page_111.QC.jpg
91693a4738540d26abf0890d9a13dd15
428618f2bb51ed0c49250d3b8656aa5d739f22a3
F20101206_AACUTM hajtas_m_Page_071.txt
084ee7357971afc4ef191188736bfdf8
b78d361d3fa5dd7a6fa46d98c822a1aa5fa3c0aa
21048 F20101206_AACUSY hajtas_m_Page_061.pro
30379589686681ac4bdfc5f3aea34182
8b487411ede52a20dbfc841ebf5c218300295cbb
5845 F20101206_AACUUB hajtas_m_Page_129thm.jpg
b6fe56f89e02e5331d2f711b9163765a
67b2344f5bc15dd1f763e4b122a373a6dba83758
F20101206_AACVXE hajtas_m_Page_046thm.jpg
c9c317fecd5fb7964c40d5b4f4929dc1
1e2bff0495d4c75401497dcf102a2b99a0a9ddb6
F20101206_AACVWP hajtas_m_Page_126.QC.jpg
101881a1ab8195e26270497768d42821
7e84139c549fdb9b993a01fe9ec40334039fb0be
1532 F20101206_AACUTN hajtas_m_Page_116.txt
62b2fca7632a49f523c458cad98533a4
762e03698cd32aa460e2eeeab5f27a893229da67
932 F20101206_AACUSZ hajtas_m_Page_133.txt
d8f403d20e300c6ea26296295d5313c2
77c10c12867d4ac67006fdc43524327a604980df
6888 F20101206_AACUUC hajtas_m_Page_105thm.jpg
3c22b9889d76e3f7eabd53f7ac4eafba
3ef2e21528b7a051ff091ec61845b52c96f0cc33
6695 F20101206_AACVXF hajtas_m_Page_036thm.jpg
ebaa96c8a619fee3d0d61958fe362ee6
a0533593ea5ea0e674d1214b52d44b7989f41cda
26805 F20101206_AACVWQ hajtas_m_Page_047.QC.jpg
5317b2c6e656fb3a215a3eff8ebb619e
3cab031da1028d4206686ee63fc18437f09b6043
29170 F20101206_AACUTO hajtas_m_Page_059.QC.jpg
c0a9ddb036f834ba04c78b40e603ef92
e3900918f4ad75ed4842db236b54b84f04fdbb01
56808 F20101206_AACUUD hajtas_m_Page_025.pro
72631658de3ead02b59076b7f53e9acf
75da2e4d74915bed696f00faf3d8656766f05c04
29416 F20101206_AACVXG hajtas_m_Page_082.QC.jpg
f3931f43b9c9e340ab8d5b71711647d4
1d8b261c4e3c7eab8b7bb22c18c5afe30fb88245
27559 F20101206_AACVWR hajtas_m_Page_035.QC.jpg
8609023349fc7629bcbca929358b458f
95865f0726cd6a427cef73fa740d6e62fa2cf8fd
18892 F20101206_AACUTP hajtas_m_Page_039.QC.jpg
92d111563e8f5589b2d6fa24895451af
14170ec8a28bc1158c69b0fe3713ea3385635c64
51670 F20101206_AACUUE hajtas_m_Page_124.pro
254ebdaac8ec4773b559536f2eae7da4
73713aeaba1ecf4ede1e044941101a2d4d4b3290
4959 F20101206_AACVXH hajtas_m_Page_006thm.jpg
341fd4d13f7cd6a27edab082eef8b35c
b8b84a4f4b38bbf93b5bed4b4eea119a4981f072
6403 F20101206_AACVWS hajtas_m_Page_048thm.jpg
42f9b896774bd90481ab3eef5228cab8
9002e288e7e4b9e2da5250ccd289692bc58b5239
29136 F20101206_AACUTQ hajtas_m_Page_030.QC.jpg
3840657d8e01e14c851eba917416cdd4
7147370d460a3d492ab977ab685407c4adfd3de9
28903 F20101206_AACUUF hajtas_m_Page_055.QC.jpg
cca0c1477d61f13aac66efa490ab49b3
ba56f18e75bc07e63f199b217c378b21aed20915
28464 F20101206_AACVXI hajtas_m_Page_044.QC.jpg
355a57b9a5190b98efbe97c2407fe520
227c3c8fdd1fbba4571df09561df54e5de7139d0
25321 F20101206_AACVWT hajtas_m_Page_092.QC.jpg
22bfe477333e45f4c67cc3d247b94506
359717041c73a9c761be00b90e0fa98f80fbdb23
54661 F20101206_AACUTR hajtas_m_Page_078.pro
afe31d294c86468802c8931211c0b89b
605093ec3ddb7e53711db2f6d5dc8c1ab2d5714d
27037 F20101206_AACVAA hajtas_m_Page_081.QC.jpg
2e541f08c65942688e7abe85ae5e7237
be52983e4dd5cad369b574d03f050130631bdd51
137074 F20101206_AACUUG hajtas_m_Page_038.jp2
05514f342fb60e5a9fc3f7104555c6a1
665a645f5fcbf4eb34f7523f12dd7a1f5e92a516
5024 F20101206_AACVXJ hajtas_m_Page_089.QC.jpg
9f27f462de2c03c2c959022d1802f80e
a1ac2adf653d0fe42a747b5a4911fa48942d65b1
1289 F20101206_AACVWU hajtas_m_Page_089thm.jpg
62bd28ddbc9c4cfba0602d3ae17e5b38
f1bcd51a9b9f65994841a7b024e36a0eccaf4d17
1543 F20101206_AACUTS hajtas_m_Page_117.txt
cdf76aa57825312e2f2a574e47b34cde
e8242938c17f1fcf6c97f4a15f6cb4d338281267
F20101206_AACVAB hajtas_m_Page_040.tif
761bb34967299e9d36b3e1799221d405
ad5f7c0a96b79c2acd03dfb85d3e6d7936e600c2
81300 F20101206_AACUUH hajtas_m_Page_085.jp2
e493154fee8cebca9ab286e622f7dadf
b64e3e985ea938c1db2d89f310067f3f4134a40d
6135 F20101206_AACVXK hajtas_m_Page_067thm.jpg
ec97a37a495dea42bb17e43de40f63d1
dada886fae934cfa3025d28decd4186ce5989dc5
6391 F20101206_AACVWV hajtas_m_Page_072thm.jpg
09650a92d959d48ee5ab8a5e0f406146
d4706c35a26c1f1bd79b6fba2d04e59ec68647c2
F20101206_AACUTT hajtas_m_Page_127.tif
cea3d28c559ce92e0146080133ad87cb
a187fd2ddc6114642943e667f8922a7e03b65d4f
6133 F20101206_AACVAC hajtas_m_Page_075thm.jpg
f2b12391226eeb8faeee62b5ab37b58f
74a1f78a2429153ea36623270c339099894d06d2
24190 F20101206_AACUUI hajtas_m_Page_132.QC.jpg
74639df88be2de237bfb4dde323094a7
3ccb71c5104ce8e160408b49629a6462f1d64804
19626 F20101206_AACVXL hajtas_m_Page_127.QC.jpg
e7f5e7b5027108f9ab38931260036e6e
900f95f0efce0796bd4673ffda07f5dd96ee6dad
6686 F20101206_AACVWW hajtas_m_Page_073thm.jpg
cfbef3081757f31261e27529a185785e
cf56882b0e6e8a9f4c4652c4dd1b5927f6b08168
20089 F20101206_AACUTU hajtas_m_Page_119.QC.jpg
be9a7c3be350edd86df0fdcced25fe1b
90b22bc193931270a4350edc35e0d97a6a77ba21
28198 F20101206_AACVAD hajtas_m_Page_109.QC.jpg
ee68d46719268f072faf790aa206e474
f6f0ddcf434d17a70914ded6eca1a264fae63911
107779 F20101206_AACUUJ hajtas_m_Page_055.jpg
76fbe1400e19c3a6dd9e53a1cf3744c2
9d5fa5222efcf3783a7d908ce494c8b862e984b7
6487 F20101206_AACVYA hajtas_m_Page_049thm.jpg
2e9fc677cd203c13e7fa27e1fdbf1be7
b4c9f257fa80ef1cbefc8876527b1c34366337fb
28503 F20101206_AACVXM hajtas_m_Page_110.QC.jpg
108cd235344b472538fbc44c3f46f9d7
859d799b5d7b9d1b4f11dbf5b99e754b23143521
6874 F20101206_AACVWX hajtas_m_Page_097thm.jpg
0df9ba2228d049fcdff2586de52c0c81
6e003127b6f622f536a479b89beea1a2332e5f6d
F20101206_AACUTV hajtas_m_Page_091.tif
1629cbc6351bdac99b1fa7459bb720b0
0432df34d79912172ae0b4915da767c8b2e89675
6677 F20101206_AACVAE hajtas_m_Page_038thm.jpg
93a4a110c8055bdaddc7062ec9297317
94e036ca75ab89e4e0de6d653f65be3a620d8674
21406 F20101206_AACUUK hajtas_m_Page_128.QC.jpg
f1f8a8b8832ade6060f76636ade90d3a
41d3db01968338f0eeb62fb9f82cfd03f46bce89
5449 F20101206_AACVYB hajtas_m_Page_130thm.jpg
16a2f6e5b1b44c8624d022a12c676b0e
6d580e619bb0e519ef61e4861882fb6738f848dc
6758 F20101206_AACVXN hajtas_m_Page_027thm.jpg
728c5e5b714973e18f2f2bf211ea21e8
7c41188c5f1228ac4c1c8f7de8376731ce3af81e
F20101206_AACVWY hajtas_m_Page_134thm.jpg
420abc03146c9f9305521b1434f6bd85
ceb7206ed05d34951ae431ead3eea1f87cc7cd59
84164 F20101206_AACUTW hajtas_m_Page_120.jp2
282e1d82fa878cde2bcecd73e1846c25
869d40ba8ff7f772188fbb1ee037a8397173daef
117756 F20101206_AACVAF hajtas_m_Page_095.jp2
750794a7fe419a51cce5d87e2ceddbe6
3d6655acd75166baef1999aef8cb5da9584263e7
107616 F20101206_AACUUL hajtas_m_Page_135.jpg
ba3274c1cfcd71eca52904486eb19101
75234d5ed2279ff8e5ce943bc4832e71842966ca
24223 F20101206_AACVYC hajtas_m_Page_131.QC.jpg
17338d58b06ecbd7f91c71fbf1e4c962
9f537c7365ec68937bb048e44e76d7752661b238
29013 F20101206_AACVXO hajtas_m_Page_103.QC.jpg
c7b7ebcc316b4c302cf3d93246146fa5
6163baf510abfc76d05ce2e31bc2715f35a1acd0
6420 F20101206_AACVWZ hajtas_m_Page_001.QC.jpg
a4fd8379cbbd5d9f87a22159ab9b76f0
aabf43b18e542ed10db6f4fc26d2084f45a36138
25288 F20101206_AACUTX hajtas_m_Page_129.QC.jpg
57459257ac0759f8e9b198ccf43b310a
5566e9375a47512a43005355ba20217383080f31
20922 F20101206_AACVAG hajtas_m_Page_006.QC.jpg
f9698a332617e3cf7b202d4def9964ec
8d13e7914e1bd969481e276df95ad177ba814d51
26752 F20101206_AACUVA hajtas_m_Page_042.QC.jpg
4bf8f66c31c8ed015be632e90b2de70a
90864f84920de3a5b3b8e1b41e2efe7edbf8c741
7029 F20101206_AACUUM hajtas_m_Page_111thm.jpg
f56cf2be581c0ae280362bc4f980b914
8f7f29182bd6a485d0e1484f1275dca93a8a5cb0







EFFECT OF SYNTACTIC STRUCTURE ON SPEECH PRODUCTION IN ADULTS WHO
STUTTER




















By

MAISA A. HAJ-TAS


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2007


































O 2007 Maisa A. Haj Tas





























To my parents: Atef Haj -Tas and Samiya Hakouz.









ACKNOWLEDGMENTS

My thanks and gratitude go to my advisor Dr. Kenneth Logan for his kind assistance

during my years at UF. I had the honor and privilege of working with Dr. Logan on several

research proj ects and clinic activities during the past six years. I learned a lot from Dr. Logan

about how to do research, how to be a good clinician, and how to excel in teaching. Most

importantly, and through his model, I learned about the qualities and value of a great mentor. I

hope that one day, I could be of similar value to my students.

I also want to thank my supervisory committee Dr. Lori Altmann, Dr. Linda Lombardino,

and Dr. Ratree Wayland for their feedback and support during the past years. Both Dr.

Lombardino and Dr. Wayland provided valuable feedback on my work and were very supportive

of my efforts. Dr. Altmann was especially helpful in assisting me to understand the details of the

methodology I used in my study. She always found time to meet with me and was very generous

in sharing and providing invaluable suggestions about how to better understand and analyze my

data.

I want to thank my friends both in the States and in Jordan for being there to share the

great times and for their support during the rough times. It was always heart warming to know

that although I was many miles away from my family, I had many brothers and sisters here to be

with whenever I needed someone to talk to or someone to listen.

I want to extend special thanks to my aunt Suad Bitamour, her husband Ibrahim Bitamour

and sons Moaied and Mahmoud for their hospitality and never ending support and

encouragement. Staying with them a few weeks every year energized my efforts and reminded

me of how wonderful it is to have family around.

Most importantly, I thank my parents Samiya Hakouz and Atef Haj -Tas and my sisters

Ghaida, Samar, Rasha, Reem, and Farah, and brothers Abdullah and Muhammad for always










being there for me. Their continuous encouragement kept me going. My mother was my biggest

supporter throughout this journey. She always showed genuine interest in whatever I was doing,

no matter how trivial that was, and always no matter how often or how late or early I called,

found the time to listen. Without her support, I would not have been able to achieve much.

I also want to thank the Department of Communication Sciences and Disorders and the

English Language Institute for supporting my study during the past years. Through their support

(both financial and moral), I was able to take classes and focus on improving my abilities as both

a scholar and a teacher.

My thanks also go to Janet Skotko at the Voice Institute and the clinicians at the

Jacksonville Speech and Hearing Center for facilitating meetings with participants for my study.

Their assistance was invaluable and was a huge factor in helping me finishing my dissertation.

Finally, I want to thank everybody who participated in my study. Although the study tasks

took on average about 80 minutes to finish, all participants were gracious, patient, and did their

best to make my efforts a success.












TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES ................ ...............9............ ....


LIST OF FIGURES .............. ...............10....


AB S TRAC T ........._._ ............ ..............._ 1 1...


CHAPTER


1 INTRODUCTION ................. ...............13.......... ......


Psycholinguistic Models of Stuttering: An overview ................. ...............14...........
The Covert Repair Hypothesis (CRH) .............. ...............14....
The Neuropsycholingui stic Theory of Stuttering ........._..... ...._... ........._.......15
The Suprasegmental Alignment Model of Stuttering ......___ ........._ ................16
The Demands and Capacities Model of Stuttering (DCM) ................ .. ......._._.......16
The Multifactorial Model of Stuttering ................. ...............17........... ...
Lingui sti c F actors and Stuttering ................. ...............19..............
Methodological Limitations................ ..............2
The Syntactic Priming Methodology .................. ... ......... ... ........... ............2
Syntactic Priming and Speakers with Typical Language Production .............................25
Syntactic Priming and Speakers with Atypical Language Production ................... .........29
The Priming Methodology and People who Stutter .............. ...............30....
Phonological Priming and PWS .............. ...............31....
Lexical/Semantic Priming and PWS .............. ...............34....
Syntactic Priming and PW S .............. ...............36....
Summary ................. ...............37.................
Research Questions............... ...............3
Question One ................. ...............39.................
Question Two .............. ...............39....
Question Three .............. ...............39....
Question Four ................. ...............39.......... ......

2 METHODS .............. ...............40....


Participants ............... ........ ...............4
Inclusion and Exclusion Criteria .............. ...............40....
Material ................. ... ........ .... .... ......... ... ..........4
Pre-Study Material Used with all Participants .............. ...............41....
Background survey (Appendix A) .............. ...............41....
W AIS vocabulary ................. ..... ........ ...............42.......
Digits forward and digits backward .............. ...............42....
Digit ordering ................. ...............43.......... .....












Pre-Study Material Used with PWS ................... ...............43..
Stuttering severity level task (Appendix A) ................. ..............................43
Self-rating scale (Appendix A) ............... .............. .. ................4
Preparation and Description of the Material Used in Syntactic Priming Task............... .44
Transitive priming sentences and pictures .............. ...............44....
Dative priming sentences and pictures ................. ...............46........... ...
Two-clause priming sentences and pictures ................. ..............................47
Filler sentences and pictures .............. ...............48....
Apparatus ................. ...............49.................
Proc edure s ................ ...............50........... ....
Data Collection ................. ...............50.................
Data Preparation for Analyses ...._.. ................. ...............51. ....
Data excluded from the final analyses .............. ...............52....
General descriptive analyses .............. ...............54...
Data preparation for speech fluency analyses ................ ................ ......... .55
Data preparation for sentence type analyses .............. ...............56....
Data preparation for SRT analyses................... ..............5
Intrajudge and Interjudge Measurement Reliability ................. ................ ........ .60

3 RE SULT S .............. ...............65....


Fluency-Related Results ................ ...............65.................
Transitive Experiment .............. ...............66....
Dative Experiment............... ...............6
Two-Clause Experiment ................. ...............70.................
Fluency Across Picture Types ................ ...............72................
Priming Analyses ................. ...............75.................
Transitive ................. ...............76.................
D ative .............. ...............76....
Two-Clause................ .. ................8
Summary of the Priming Analyses ................. ...............81........... ...
SRT Analyses .............. ...............8 1....

4 DI SCUS SSION ................. ...............92................


The Effect of Syntactic Structure on Fluency............... ...............93
The Effect of Picture Type on Fluency ................. .. .......... ...............98....
The Effect of Syntactic Structure on Speech Reaction Time .............. ....................9
The Effect of Syntactic Priming on Response Type................ ...............105.
Differences Between The Two Groups On the Prestudy Tasks ................. ............... .....110
Conclusion ................. ...............111......... ......


APPENDIX


A PRE STUDY TE STS ................. ................. 113........ ....


B PRIMING AND FILLER SENTENCES AND PICTURES ................ .......................120











LIST OF REFERENCES ................. ...............134................

BIOGRAPHICAL SKETCH ................. ...............138......... ......


































































8










LIST OF TABLES


Table page

2-1 Participant Demographics and Performance on the Pre-study Language and Memory
Tests ......._ ......... ...............64.......










LIST OF FIGURES


Figure page

2-1 Mean and standard deviation for the number of responses produced across the
examined sentence types (Transitive, Dative, and Two-Clause) in addition to
responses that exhibited syntactic structures other than the examined types (Other)........62

2-2 Mean and standard deviation for number of words produced for (a) Transitive
responses in the Transitive experiment, (b) Dative responses in the Dative
experiment, and (c) Two-Clause responses in the Two-Clause experiment. ................... ..62

2-3 Mean and standard deviation for the percentage of sentences produced when the
sentence was entirely fluent, when the first noun phrase in the sentence was fluent,
and when the first word in the first noun phrase in the sentence was fluent .....................63

3-1 Mean number and standard error for 'repetitions and prolongations' (Rep.& Pro.)
and 'other' types of disfluencies (Other) when the priming sentences were (a)
Transitive, (b) Dative, and (c) Two-Clause. ............. ...............86.....

3-2 Mean number and standard error for 'repetitions and prolongations' (Rep.& Pro.)
and 'other' types of disfluencies (Other) in responses when the priming pictures were
(a) Transitive (b) Dative, and (c) Two-Clause ................. ...............88........... ..

3-3 Mean percentage and standard error for responses when the priming sentences were
(a) Transitive, (b) Dative, and (c) Two-Clause ................. ...............89........... ..

3-4 Mean and standard error for speech reaction time (SRT) when the responses were(a)
Transitive in the Transitive experiment, (b) Dative in the Dative experiment, and (c)
Two-Clause in the Two-Clause experiment. ............. ...............91.....

3-5 Mean and standard deviation for number of words for (a) Transitive responses in the
Transitive experiment, (b) Dative responses in the Dative experiment, and (c) Two-
Clause responses in the Two-Clause experiment ................. .............. ......... .....91









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECT OF SYNTACTIC STRUCTURE ON SPEECH PRODUCTION IN ADULTS WHO
STUTTER

By

Maisa A. Haj-Tas

December, 2007

Chair: Kenneth J. Logan
Major: Communication Sciences and Disorders

This study examined (a) the effect of syntactic structure on speech fluency and speech

reaction time (SRT) in adults who stutter, (b) the effect of syntactic priming on fluency in those

participants. Fourteen persons who stutter (PWS) and fourteen persons who do not stutter

(PWN\S) participated in the study. A sentence structure priming paradigm was used to elicit

sentences of varying syntactic forms (i.e., transitive, dative, and two-clause).

The results of the fluency analyses indicated that (a) PWS produced significantly more

'repetitions and prolongations' per response than PWNS, (b) all participants produced a

comparable number of 'other' types of disfluencies (e.g., interjections and revisions), (c) all

participants seemed to produce more fluent responses following transitive pictures and fewer

fluent responses following two-clause pictures.

The results of the priming analyses indicated that (a) the presence of passive primes did not

significantly increase the probability of using passives in the response, (b) the structure of the

dative primes affected the probability of using those structures in the response differently

between the study groups. Specifically, the PWS produced significantly more prepositional-

dative responses following prepositional dative primes than following obj ect-complement

primes; however, no significant differences among the effect were observed in the PWNS. This









finding was taken to suggest that priming dative sentence forms for the PWNS might be more

driven by the structural aspects of the available options and for the PWS by the thematic aspects,

and (c) the presence of the embedded primes did not significantly increase the probability of

using embedded structures in the responses.

The results of the SRT analyses indicated (a) no significant differences between the two

groups in SRT, (b) that the syntactic structure of the response did not affect SRT differently

between the two groups, and (c) all participants produced transitive responses at a significantly

shorter SRT than dative responses, and dative responses at a significantly shorter SRT than two-

clause responses. The results of the SRT were taken to suggest that increasing the syntactic

complexity of the responses may influence the time both the PWS and the PWNS may need to

generate such responses.









CHAPTER 1
INTTRODUCTION

The relationship between speech formulation processes and stuttering has been central to

numerous hypotheses and investigations for several decades. A number of models of stuttering

suggest stuttering is a problem that results from difficulties arising during speech formulation,

that is, prior to speech production (e.g., Kolk & Postma, 1997; Perkins, Kent, & Curlee, 1991;

Starkweather, 1987). Results of numerous studies also provide evidence that increased linguistic,

temporal, and cognitive loads seem to affect the speed and fluency of speech production in

persons who stutter (PWS) more so than it does in persons who do not stutter (PWNS) (e.g.,

Bernstein Ratner & Sih, 1986; Bosshardt, 1993; Bosshardt & Fransen, 1996; Cuadrado &

Weber-Fox, 2003; Logan & Conture, 1995, 1997; Logan, 2001, 2003; Silverman & Ratner,

1997; Yaruss, 1999). Despite all this evidence, there are still numerous unanswered questions

about the role that speech formulation plays in the disfluencies that characterize stuttered speech.

The present study was conducted to examine the role that syntax might play in speech production

in general, and speech fluency and speech timing in specific in PWS.

The plan of this chapter is as follows: Section I provides an overview of a number of

psycholinguistic models that offer different perspectives about the cause of stuttering. Section II

provides a review of empirical evidence that supports the effect of certain linguistic variables,

namely syntactic encoding, on the efficiency and timing of speech formulation processes in

PWS. Section III presents a number of methodological limitations associated with previous

research that has explored the effect of syntax on speech production processes in PWS. Section

IV reviews a methodology (i.e., syntactic priming) that has been used to examine speech

formulation processes in both speakers with typical language and speakers with atypical

language functioning, and which has only been used infrequently to examine speech formulation










processes in PWS. Finally, section V presents the rationale for using the syntactic priming

methodology to examine the syntactic encoding abilities in and between PWS and PWNS.

Psycholinguistic Models of Stuttering: An overview

Numerous models of speech production suggest that different aspects of speech

formulation might be involved in the production of disfluencies by PWS (for a review see

Bernstein Ratner, 1997; Conture 2001). Such models seem to have a common underlying claim

that disfluencies produced by PWS result from disruptions in linguistic planning prior to speech

production (Karniol, 1995; Kolk & Postma, 1997; Perkins, Kent, & Curlee, 1991; Smith &

Kelly, 1997, Starkweather, 1987). In this section, I provide an overview of the five best-

developed models of speech production that researchers have used in their investigations of

linguistic planning abilities of PWS.

The Covert Repair Hypothesis (CRH)

Developed by Postma and Kolk (1993), this theory proposes that fluency breakdowns in

stuttering result from difficulties that arise prior to speaking in general and at the phonological

encoding level in specific. The CRH suggests that although the self-repair process and speech

production monitoring skills in PWS are similar to those in typical speakers, the processes

underlying the selection of phonemes or insertion of the selected phonemes into speech plans are

slower or delayed in PWS compared to PWNS. This "slowness" in phoneme activation or

encoding results in particular phonemes being in competition for selection with other phonemes

for a time frame longer than normal, which in turn increases the chance of phoneme selection

errors if the speaker attempts to commence speaking before the phoneme selection process is

complete.

According to the CRH, stuttering arises when PWS detect such selection errors and try to

correct them while speaking at a rate that is faster than their impaired phonological encoding









mechanisms allow them to proceed. Within this view, the repetitions and prolongations observed

in the speech of PWS are a result of attempts by those speakers to self repair the errors before

they are spoken (i.e., covertly repair those errors). In Conture' s (2001) review of the CRH, he

explained that although this hypothesis has several components that make it appealing (e.g.,

"attempting to account for speech disfluencies that arise during conversational speech"), the

CRH does not account for other levels of speech formulation such as lexical/semantic and

syntactic encoding. Thus, as Conture suggested, the CRH "may have limited ability to handle the

sort of individual variations that other models so aptly attempt to describe." (p.37)

The Neuropsycholinguistic Theory of Stuttering

Developed by Perkins, Kent, and Curlee (1991), this theory suggests that stuttering results

from dyssynchrony in the integration of linguistic units into their planned speech frames.

Specifically, the theory suggests that during speech formulation, stuttering occurs if the

generation of sounds or "fillers" and the integration of those fillers into syllable frames (or slots)

become dyssynchronous. Perkins et al. (1991) explained that hesitations during speech are

expected to arise as a result of interactions between uncertainties about the planned linguistic

unit and the need for the speaker to "begin, continue, or accelerate an utterance" (p.734). Within

such a framework, nonstuttered disfluencies such as revisions and interj sections could occur when

(a) a speaker is under time pressure and the cause of disfluency such as the speaker' s uncertainty

about the linguistic characteristics of the target is known, or (b) when the speaker does not know

the cause of uncertainty but is not under time pressure. On the other hand, stutter-like

disfluencies (e.g., repetitions and audible and inaudible prolongations of sounds and syllables)

may occur under the combination of two "primary conditions" (a) when disfluencies could arise

from a variety of causes and the speaker is not certain about which cause is contributing to them,

and (b) if the speaker is under real or perceived pressure to continue speaking when such









disruptions occur. Within this framework, Perkins et al. (1991) define stuttering as "disruption of

speech that is experienced by the speaker as a loss of control" (p.734). Thus, time pressure is a

central component in stuttering within this theory's framework.

The Suprasegmental Alignment Model of Stuttering

Developed by Karniol (1995), this model suggests that stuttering results from problems at

the speech formulation level. Unlike Kolk et al. (1993) and Perkins et al.'s (1991) models of

stuttering, Karniol suggested that stuttering arises from difficulties at the sentence level. The

suprasegmental alignment model proposes that disfluencies produced by both PWS and PWNS

result from the attempt of the speakers to revise sentence plans during online sentence

production. Within such a perspective, Kamiol explained that suprasegmental features of a

sentence such as rhythm and stress "are largely determined prior to utterance initiation and are

expressed through changes in muscle movements" (p. 111). Additionally, and as Kamiol

explained, although suprasegmental features are prepared before the sentence is uttered, speakers

often change their speech plans online. Such changes most often occur because speakers may

initiate speech before the utterance planning process (especially that for the verb phrase) has

been completed. If problems arise during speaking and the utterance cannot be completed as

originally planned, speakers may change those plans. Kamiol suggests that both the latencies and

hesitations observed in speech production reflect such sentence plan changes. In some sense,

Karniol's model is similar to Perkins et al.'s model in that both propose that increased demands

(such as the need to continue talking while attempting to correct selection errors under time

pressure) are important contributors to the transformation of disfluencies into "stuttering events".

The Demands and Capacities Model of Stuttering (DCM)

This model provides yet another view of the processes that lead to stuttered speech (e.g.

Starkweather, 1987). Similar to Perkins et al. neuropsycholingui stic model, the DCM suggests









that increased demands affect speech production in PWS. The central claim in the DCM model is

that stuttering (specifically in children) may result from demands within or between domains that

exceed the capacities of the speaker to produce speech fluently (for a review and critique of the

model, see Bernstein Ratner, 1997; Conture, 2001; Manning, 2000; Siegel, 2000, Yaruss, 1999,

2000). Much of the available critique of this model argues that although it is very appealing, the

nature and level of demands are not well defined in the model. Some authors, however, provide

their own interpretation of the claims of the DMC and suggest examples that may help better

understand the nature of such demands. For example, in his review of this model, Conture (2001)

provided one example to show how the assumptions of this model work during speech. In his

example, Conture explained that although a speaker may have typical (or above average)

capacities for generating utterances that are long or grammatically complex, these capacities

might be challenged under some conditions, such as when the speaker attempts to rapidly

produce long or grammatically complex utterances that may contain complex articulatory

adjustments. Within the framework of the DCM, such time pressure demands may exceed the

speaker' s ability to make the speech motor adjustments needed to accurately and efficiently

produce such utterances, and this in turn increases the probability of stuttering.

The Multifactorial Model of Stuttering

Developed by Smith and Kelly (1997), this model provides yet another view of the factors

underlying stuttering and suggests that stuttering results from factors that interact in a complex,

nonlinear, and dynamic manner. Smith and Kelly explained that stuttered events should not be

viewed as static behaviors that occur in isolation at discrete points in time. Rather, stuttered

events should be examined within a multifactorial framework in which interactions among

various important factors are analyzed. The model also suggests that definitions and descriptions

of stuttering need to take into account the reality that the nature of the interactions among factors









will likely vary from one context to another and will also vary between speakers. The Smith et

al. model seems to be similar to the other models reviewed in this section in the sense that all of

the previously reviewed models suggest that, in one way or another, stuttered speech could result

from interactions between two or more factors. The appeal of Smith and Kelly's model is in the

fact that it does not confine the cause of stuttering to just one factor and that it considers the

variability of stuttering under different conditions. Also, and as Conture (2001) suggests in his

review of this model, "The Smith and Kelly model does a nice j ob of pointing out that individual

variations among different variables may mean that we should be talking about causes rather

than the cause of stuttering." (p.33).

Although the models reviewed above adopt different approaches in their attempt to explain

the factors underlying stuttering, in one way or another, they all seem to suggest that stuttering

might result from difficulties that arise during language formulating and prior to speech

production. As Conture (2001) suggested, such models may lead us to conclude that some of the

factors underlying stuttering "might be found between thought and motor execution of that

thought" (p. 354). Although other factors such as motor execution have been implicated in

stuttering, Conture suggested that the components of speech production processes above the

motor execution level such as semantic, phonological, and syntactic components are "fast

enough and creative enough to account for what we know about stuttering" (p.354). Indeed,

numerous studies have examined the relationship between speech formulation processes and

stuttering and results of several of those studies suggest that such a relationship might exist. The

following section provides a review of those studies and the main findings they reported. In

addition, the section provides evidence from studies with both children and adults who stutter









about the effect of different linguistic factors in general and the effect of grammatical or

syntactic complexity in specific on speech fluency in those speakers.

Linguistic Factors and Stuttering

Empirical evidence from numerous studies has consistently shown that mechanisms

underlying language formulation processes such as phonological, semantic, and syntactic

encoding may operate atypically in PWS compared to their non stuttering peers (Bernstein

Ratner, 1997; Bosshardt, 1993; Bosshardt & Fransen, 1996; Cuadrado & Weber-Fox, 2003; Kolk

& Postma, 1997; Wijnen & Boers, 1994; Weber-Fox, Spencer, Spruil, and Smith, 2004).

Although on one hand results of several such studies suggest that PWS on average perform

within the normal range on static, offline measures of linguistic abilities (e.g., Nippold, 1990;

Watkins, Yairi, & Ambsrose, 1999), on the other hand, results of numerous studies suggest that

the combination of a "fragile" linguistic system and a variety of linguistic and cognitive factors

may result in subtle differences between PWS and PWNS during online speech processing

(Bernstein Ratner, 1997; Bosshardt, 1993; Bosshardt & Fransen, 1996; Cuadrado & Weber-Fox,

2003; Weber-Fox, Spencer, Spruil, & Smith, 2004).

Although different linguistic processes involved in speech production in PWS have been

examined thus far, the contribution of grammatical or syntactic complexity has received most of

the attention during the past years. Specifically, many investigations have been conducted to

examine the nature of syntactic events that contribute to stuttering and the effect of manipulating

such events on speech fluency and accuracy in PWS. In the following section, a brief review will

be provided of the main findings of such investigations.

Research on the location and frequency of disfluencies within utterances produced by

speakers who stutter indicate that disfluent events do not occur randomly during speech.

Specifically, and as Peters and Guitar (1991) explained, "Stuttering occurs more often at points









in the utterance that can be described in linguistic terms." (p. 117). For example, results with

PWS have consistently shown that stuttering tends to be observed more often on (a) sounds in

the word-initial position rather than sounds in other word positions, (b) consonants rather than

vowels, (c) multisyllable words rather than monosyllable words, (d) content words (e.g., nouns,

verbs, adj ectives, adverbs) rather than function words (e.g., articles, prepositions, and pronouns),

(e) stressed syllables rather than unstressed syllables, (f) low frequency words rather than high

frequency words, and (f) words occurring at or near the sentence initial position rather than

sentence final position (for more information, see reviews in Bloodstein, 1995; Bernstein Ratner,

1997; Peters & Guitar, 1991).

Results of numerous studies have also shown that in children who stutter (CWS), stuttering

is more likely to occur within utterances that are long and/or syntactically complex than

utterances that are short and/or syntactically simple (e.g., Bernstein Ratner and Sih, 1987;

Gaines, Runyan, & Meyers, 1991; Gordon, Luper, & Peterson, 1981; Logan & Conture, 1995;

1997; Wall, Starkweather, & Cairns, 1981; Yaruss, 1999). In general, these results have been

interpreted within theoretical frameworks such as the ones reviewed earlier in this chapter. For

example, within the framework of the DCM, increased utterance length and syntactic complexity

may be viewed as demands that exceed the child's capacity to produce such utterances fluently

(e.g., Logan & Conture, 1997).

Although, as mentioned above, linguistic complexity has been shown to affect fluency in

CWS, investigations of this effect in older speakers who stutter provide different results (e.g.,

Logan, 2001; Silverman & Bernstein Ratner, 1997). For example, Silverman and Ratner (1997)

examined the relationship between syntactic complexity and fluency in adolescents who stutter

and adolescents who do not stutter. The participants heard and then repeated utterances which









featured one of three levels of grammatical complexity (i.e., "Wh" questions, right-embedded

relative sentences, and center embedded relative sentences). The authors measured the effect of

utterance complexity on the accuracy and fluency of sentence imitation by the two groups. The

results of the study indicated that increased grammatical complexity affected both speaker

groups in a similar manner. Specifically, as the grammatical complexity of the target utterance

increased, the percentage of disfluencies and inaccurate responses also increased in both groups.

Based on the results of their study, Silverman et al. (1997) concluded that although some

syntactic structures continue to be challenging for both adolescents who stutter and adolescents

who do not stutter (as observed in the comparable increase in speech disfluencies in both groups

in the more syntactically challenging sentences), the effect of syntactic complexity appears to

diminish as speakers grow older and their linguistic competency improves.

In a similar investigation, Logan (2001) studied the effect of increased syntactic

complexity on speech fluency and timing in younger (i.e., adolescents who stutter) and older

PWS as compared across two speaking contexts: (a) a conversational speech task during which

the participants generated utterances during a structured conversation, and (b) a sentence reading

task that was performed within a reaction time paradigm. In the sentence reading task, the

participants read and rehearsed a series of sentences in which the subj ect noun phrase differed in

syntactic complexity. In detail, Logan prepared a base level sentence (e.g., The cake was served

at the dinner) and then generated four sentence versions from the base sentence by changing the

structure in the "subj ect constituent" of the sentence. As Logan explained, in sentence version

one, a Determiner + Adj ective + Adj ective + Noun subj ect constituent was used (e.g., The rich

and tasty cake was served at the dinner). In sentence version two, the subj ect constituent was

Determiner + Noun + Prepositional Phrase structure (e.g., The cake from Bill's oven was served









at the dinner). In sentence version three, the subj ect constituent was Determiner + Noun +

Relative Clause (The cake that was fancy was served at the dinner.). Finally, in sentence version

four, the subj ect constituent was Determiner + Noun (e.g., The cake was served at the dinner for

the mayor) (For more details, see Logan, 2001, pp. 92-93). Participants were presented with and

asked to reproduce the sentences initiating them as promptly as possible following the

presentation of a cue (i.e., 1 KHz a pure tone). Consistent with the findings reported by

Silverman et al. (1997), Logan reported that speech fluency is comparable between the two

groups when producing sentences of varying syntactic complexity. In other words, different

levels of syntactic complexity seemed to similarly affect speech fluency in adolescents and

adults who stutter. Logan also reported that the means for the different disfluency types

examined in the study were similar for the two groups. Based on these finding, Logan concluded

that younger PWS seem to control their level of speech fluency and manage sentences of varying

grammatical complexity in a manner different from that used by older PWS. Based on those

results, Logan suggested, in agreement with Silverman et al., that the factors that underlie

stuttering in children might not persist beyond the childhood years.

Methodological Limitations

Although the results from Silverman et al. (1997), and Logan (2001) may appear to argue

against a role for syntactic complexity in stuttering in adolescents and adults who stutter, the

authors of both studies suggested that using speech elicitation paradigms other than the ones used

in their studies (i.e., paradigms other than sentence repetition) might provide information about

subtle syntactic effects that exist in the speech production process of PWS. For instance, Logan

noted that, in his study, syntactic complexity was defined within a transformational grammar

framework. He also suggested that alternate frameworks of complexity, such as those based upon

"memory cost" models, might yield different results.









Indeed, a number of alternative paradigms examining the relationship between syntactic

complexity and stuttering have provided a different view of this relationship. Such alternative

paradigms include using dual-task activities (e.g., Bosshardt, Ballmer, & De Nil, 2002), online

grammaticality judgments (e.g., Cuadrado & Weber-Fox, 2003), and speech reaction time (SRT)

(e.g., Logan, 2003). Although the three above paradigms are different in terms of their

procedural details, the motivation underlying their use seems to be similar in that they all

measure the effect of syntactic complexity on speech production in individuals who stutter under

conditions of increased cognitive and time loads during online investigations of the speech

production process. As suggested by Cuadrado and Webber-Fox (2003), "Additional cognitive

loads (e.g., in a dual-task paradigm) have been shown to accentuate differences between NS [non

stutterers] and IWS [individuals who stutter] in planning of linguistic units in speech

production." (p. 962). The results of such studies have consistently shown that although there is

an overlap in the response fluency between PWS and PWNS, increased cognitive loads and

interfering "attention-demanding" tasks seem to have a more pronounced effect on the efficiency

of retrieval and or encoding of linguistic units during sentence formulation in PWS. For example,

results of several studies revealed that under dual tasks (compared to single tasks), PWS require

more time for sentence generation and articulation and produce more speech disfluencies than

PWNS (Bosshardt, Ballmer, & De Nil, 2002; Caruso, Chodzko-Zajko, Bidinger, & Sommers,

1994). Similarly, results of other studies revealed that under more demanding 'online tasks' (i.e.,

tasks that are used to measure the effects of a given variable on speech production during

ongoing speech processing) versus offlinee tasks' (i.e., tasks that are used to measure the effects

of a given variable on speech production after or at the end-point of speech processing), the

accuracy of grammaticality judgment in general and syntactic processing in specific in PWS are









most affected by the increased length and syntactic complexity of the target sentence during

online tasks as compared to offline tasks (e.g., Cuadrado & Weber-Fox, 2003).

As mentioned above, although syntactic complexity has repeatedly been shown to

contribute to breakdowns in fluency in CWS, investigations of this linguistic phenomenon in

older PWS seem to provide mixed results. For example, Silverman et al. (1997) and Logan

(2001) suggest that the effect of syntactic complexity on fluency might be minimal in older

speakers who stutter, while Bosshard (1993) and Cuadrado et al. (2003) suggest that this effect

can be observed under certain cognitive demands. The difference between the results offered by

the these two study groups could be attributed to several factors among which are the

methodological differences. The later group of studies has examined the effect of syntactic

complexity on speech formulation and production abilities of PWS when a concurrent or

secondary task was used. The results of those studies show a significant effect of linguistic

phenomena in PWS. Thus, methodologies that used concurrent tasks seem to add to the cognitive

loads during those experiments and that in turn offered a glimpse into the dynamics of this

interaction. The following section provides an overview of an alternative online, and thus more

demanding methodology- syntactic priming. This methodology has been used for several

decades in psycholinguistics research to examine speech formulation processes in both typical

speakers and speakers with language delays. As will be shown below, and in agreement with

several other researchers, I believe that syntactic priming is a potentially informative

methodology for examining syntactic encoding abilities of PWS.

The Syntactic Priming Methodology

Many observations of natural language have reported that speech is "highly repetitious"

(e.g., Levelt & Kelter 1982; Bock, 1986). For more than two decades, researchers have

successfully shown that the repetitious nature of language can be induced under experimental









conditions by using the priming methodology. The following section will provide an overview of

the priming methodology in general, and the syntactic priming methodology in specific. In

addition, the following section will provide an overview of the studies that have used the priming

methodology to examine the speech formulation processes in speakers who stutter, and

particularly, the syntactic processing abilities in this population. A review of phonological and

lexical/semantic priming studies will be included for the sake of establishing the feasibility of

this methodology in exploring the speech formulation and production processes in PWS.

Syntactic Priming and Speakers with Typical Language Production

It is well documented that, during conversations, speakers tend to repeat grammatical

structures that either they or their interlocutors produced earlier in those conversations (for a

review, see Bock, 1986; Pickering & Branigan, 1999; Branigan, Pickering, McLean, & Stewart,

2006; Smith & Wheeldon, 2001). As Smith and Wheeldon (2001) explained, repetition in natural

language "seems to affect all aspects of the grammatical structure of language such as syntactic

frames, formulaic language, individual lexemes or phases, or the lexicon itself'(p.124). Bock

(1986) was among the first researchers to show that the repetition of one particular grammatical

structure that has been observed in natural language can also be observed in experimental

settings. This phenomenon has been termed, variably, structural persistence, syntactic

persistence, sentence structure priming or syntactic priming. The term 'syntactic priming' will be

used in the present study.

Several hypotheses have been proposed to explain the syntactic priming phenomenon. In

her pioneering examination of this phenomenon, Bock explained that her studies were designed

to explore 'The syntactic activation hypothesis,' which examines "grammatical patterns in

speech result from the application of cognitive realizations of syntactic rules or structure

heuristics, perhaps via procedural representations of grammatical structures" (p. 358). Bock










suggested that one consequence of such an application could be a strengthened or elevated level

of activation of certain rules or structures, which result in the repetition of a given syntactic

structure. To explore this hypothesis, Bock prepared sentences that contained syntactic structures

reported in earlier observations to be associated with the persistence (repetitious) effect such as

dative sentence structures (e.g., The boy gave the girl a valentine) and passive sentence

structures (e.g., The man was stung by a bee).

During the experiment, Bock presented participants with sentences and pictures that were

transitive (i.e., the main in those sentences and pictures requires an obj ect as in 'The fireman

rescued the baby') and dative (i.e., the main verb in those sentences and pictures requires two

obj ects as in 'The boy ") gave the girl(2) a valentine'). The participants first heard then repeated

sentences such as the transitive sentence 'One of the fans punched the referee.' Next they saw a

picture such as that of a tornado destroying a bamn and were asked to briefly describe what was

happening in that picture in one complete sentence without using any pronouns. The design of

the study was such that each of the transitive target pictures could be described by either an

active sentence (e.g., A tornado destroyed the bamn) or a corresponding passive (e.g., The bamn

was destroyed by a tornado). Similarly, each of the dative pictures (e.g., a boy giving a valentine

card to a girl) could be described either by a prepositional dative sentence (e.g., The boy gave a

valentine to the girl) or a double obj ect dative sentence (e.g., The boy gave the girl a valentine).

The results of the study showed that the presentation of a certain syntactic structure (i.e.,

syntactic prime) prior to a target picture increased the probability of that structure being used to

describe the target picture. For example, participants seemed to show an increased use of active

structures to describe a picture when the priming sentence they repeated was of an active

structure. In a series of follow-up experiments, Bock and colleagues provided evidence that the










generated syntactic structures were not a result of conceptual characteristics of the sentences, and

that the priming effect was observed regardless of the content of those priming sentences (Bock,

1986; Bock and Loebell, 1990). Based on the results of those studies, Bock and colleagues

concluded that the structural persistence (i.e., priming) effect observed in their experiments was

actually a result of activation of syntactic processes, which in their opinion, supported the above

mentioned syntactic activation hypothesis. Bock (1986) also suggested that "although processes

such as priming may limit the flexibility of syntax in spontaneous real time use of language,

perhaps contributing to differences in syntactic diversity between planned and extemporaneous

speech," (p.379), such a process could also have "an adaptive function." Bock further explained

that the expression of unplanned messages during speech "can create problems that lead to

hesitations, errors, and other disruptions (p.379). As suggested by Bock, reusing structures that

were available in an earlier sentence (within a context similar to the one used in the study) could

contribute to fluency since "it employs procedures that are already activated" and this in turn

may ease the demands of generating the target message.

These conclusions were tested more recently in a series of six experiments conducted by

Smith and Wheeldon (2001). The experiments examined the effect of syntactic priming on

speech initiation times (SIT), among other variables, in typical adult speakers. The authors

suggested that if syntactic persistence results from reusing already activated procedures,

syntactic priming may positively affect temporal factors in speech production by reducing the

time speakers need to initiate utterances. The first experiment reported by Smith et al. (2001)

was designed to investigate the hypothesis that syntactic persistence reduces the time dedicated

to syntactic planning (as observed in reduced speech onset latencies or a shorter SIT). The

material of the study consisted of a variety of moving pictures. In the 'syntactically related'









condition and as described by Smith et al. the prime and target trial sentences were matched

both in terms of the grammatical role of the phrases they assigned nouns to (i.e., a subj ect

phrase) and in terms of the complexity of the internal structure of the phrase they assigned nouns

to (i.e., a conj oined noun phrase."(p. 130). An example of a target phrase used in the study is

'The spoon and the car move up,' and a corresponding syntactically related sentence is 'The eye

and the fish move apart.' In the "syntactically unrelated" condition, Smith and Wheeldon (2001)

explained that "the prime trials matched the target trials in terms of the grammatical role of the

phrases they assigned nouns to but not in terms of the complexity of the internal structure of the

phrase they assigned nouns to (i.e., a simple noun phrase rather than a conj oined noun phrase).

(p. 130). An example of a syntactically unrelated priming sentence used in the study is 'The eye

moves up and the fish moves down.'

During the experiment, participants described black and white line drawings of obj ects

moving in different directions on a computer screen. Prior to the experiment, the participants

were instructed to describe the picture movements from left to right using specific sentence

types. Speech production latencies were measured to determine whether there were significant

differences in speech latencies between the syntactically related and syntactically unrelated

conditions. The results of the study revealed that target sentences that were preceded by

syntactically related primes were produced at significantly shorter (55 ms) speech latency than

target sentences that were preceded by unrelated syntactic primes. Based on the results of the

study, Smith and Wheeldon concluded that "Clearly, such a result provides the first evidence

compatible with the hypothesis that syntactic persistence benefits speakers by reducing the

processing costs incurred by syntactic structure generation" (p.138).









As observed in both Bock (1986) and Smith and Wheeldon (2001) experiments, syntactic

priming seems to be a reliable methodology for examining sentence structure generation and

production processes in typical speakers. In addition, the presentation of syntactically related

primes prior to targets seems to facilitate sentence generation processes as measured by the

reduced speech latencies under the related syntactic prime condition. In an above-presented

explanation of how the syntactic effect works, Bock had suggested that although syntactic

persistence might "limit the flexibility of syntax in spontaneous speech" it might also have "an

adaptive function" in the sense that the expression of unplanned messages during speech "can

create problems that lead to hesitations, errors, and other disruptions" (p.379). Both Bock (1986)

and Smith et al. (2001) also seem to agree that message generation under some conditions might

impose additional cognitive loads that may affect the timing and fluency of the uttered sentence.

Since syntactic priming has been shown to enhance fluency during real time tasks in

typical speakers, numerous studies have examined this phenomenon and whether similar results

can be observed with atypical speakers. The following section provides results of such studies.

Syntactic Priming and Speakers with Atypical Language Production

In addition to its use with speakers exhibiting typical language abilities, the priming

methodology has also been used to examine syntactic processing abilities in speakers with

atypical language production. For example, the syntactic priming paradigm has been used to

examine difficulties in grammatical processing in aphasic patients (e.g., Hartsuiker & Kolk,

1998; de Roo, Kolk, & Hofstede, 2003), patients diagnosed with dyslexia (Faust, Silber, &

Kaniel, 2001), and patients diagnosed with schizophrenia (e.g., Besche, Passerieux, Segui,

Sarfati, Laurent, & Hardy-Bayle, 1997).

For example, Hartsuiker et al.'s 1998) study revealed that patients with Broca' s aphasia

showed stronger syntactic priming effects than the typical control speakers, and that although the










speech of such patients is usually characterized by a reduced complexity of syntactic structure,

these patients produced relatively complex sentence structures (such as passives) when those

structures were primed. Hartsuiker et al. (1998) explained that some hypotheses suggest that

patients with Broca' s aphasia might suffer from reduction in their capacity for generating

grammatical structures, especially complex ones, which may result from either slowness or

"reduced maintenance"/"fast decay" in the generation of grammatical information. Within this

view, and based on the results of their study, Hartsuiker et al. provided a number of conclusions

among which was that limitations in computational resources in patients with Broca' s aphasia

can be overcome by using facilitatory processes such as syntactic priming.

Besche et al. (1997) used syntactic priming to examine the hypothesis that some types of

schizophrenic disorders might have the "disorganization of cognitive" activities as an underlying

cause and particularly disorganization of semantic and syntactic relations during the processing

of lexical information. The results of the study revealed that the performance of the

schizophrenic patients pertaining to the processing of syntactic information was identical to that

of the typical controls. In addition, results revealed that both groups showed a significant

syntactic priming effect as observed in the form of reduced SIT during a decision making task.

As can be concluded from the two examples above, syntactic priming is a methodology

that can be helpful in examining syntactic formulation and production abilities in atypical

populations as well as typical speakers. The following section provides an overview of a number

of studies that have used the priming methodology to examine speech processing abilities of

PWS.

The Priming Methodology and People who Stutter

As mentioned above, both empirical evidence and theoretical explanations suggest that

stuttering might result from difficulties in generating or encoding syntactic units. In addition, a









number of hypotheses suggest that speakers who stutter may exhibit limited capacities to

generate fluent speech under conditions of increased linguistic or cognitive demands (DCM; e.g.,

Starkweather, 1987). If this is the case, then the facilitatory effect of syntactic priming that has

been shown in several clinical populations who exhibit limitations in speech generation

capacities (e.g., patients with Broca' s aphasia) may also enhance speech fluency in PWS.

Although the priming methodology has been used in studies of speech production with

both typical and clinical populations as shown in the previous section, its application to stuttering

is still in its infancy. Most of the studies that have used priming to examine speech formulation

in PWS have used priming to examine the time course of phonological and lexical/semantic

encoding in children and PWS (e.g., Burger & Wijnen, 1999; Hartfield & Conture, 2006;

Melnick, Conture, & Ohde, 2003; Pellowski & Conture, 2005; Wijnen & Boers, 1994). To my

knowledge, only one study thus far used priming to examine the syntactic processing abilities of

speakers who stutter (Anderson & Conture, 2004). Although as mentioned syntactic priming is

the scope of the present study only covers syntactic priming, I believe that is it worth presenting

studies that used other priming methodologies such as phonological and lexical/semantic priming

in persons who stutter. Presenting such studies will help establish that the priming methodology

can be reliably used to examine speech formulation and production abilities of this population.

The following section first provides an overview of studies that have used the priming

methodology with PWS and the main results of those studies, and then provides the rationale for

using syntactic priming to examine syntactic processing abilities of PWS.

Phonological Priming and PWS

Wij nen and Boers (1994) were the first researchers to use the priming methodology to

examine phonologic encoding abilities of PWS. The study was designed to test the hypothesis

that stuttering results from problems in phonological encoding and particularly the encoding of










syllable rime ("the syllable constituent capturing the vowel and any Einal consonants" (p.5)).

Wijnen and Boers explained that "It is assumed that priming increases the activation level of the

involved phonological segments) so that incorporation [of those segments] in the articulatory

plan is facilitated." (p.6). The participants in the study were nine adults who stutter and nine

adults who do not stutter (all native speakers of Dutch). Two phonological priming conditions

were used: priming the initial consonant (C-Block), and priming the initial consonant and

subsequent vowel (CV-Block). A cue word was visually presented and the participants had to

utter a response word (out of a set of five words) that was assigned to the cue word and which

the participants learned prior to the online experiment. There were Hyve blocks of Hyve-word sets

that were either homogeneous (all words shared the same consonant or shared the same

consonant and vowel) or heterogeneous (words were a combination from each of the

homogeneous sets). For example, in the homogeneous condition of a C-Block, the cue word was

"vork" the corresponding response words were "Lepel", "Lila", "loeder", "larie", and "Luir".

Speech reaction time was measured in each of the priming conditions. The results of the study

indicated that SRT in the C-Block was significantly faster during the homogeneous condition

(606 ms) than it was in the heterogeneous condition (639 ms), indicating a larger priming effect

during the former condition (although, as the authors reported in a follow up study presented

below, this effect was only observed in only four out of the nine study participants). The authors

also reported that in most participants who stutter, reduction of SRT occurred only when the

words shared both the consonant and vowel. Based on the results of the study, Wijnen and Boers

concluded that there is a difference in phonological encoding abilities between PWS and PWNS,

and that the results indicate that the encoding of stressed vowel at the non-initial parts of the

syllable might be delayed PWS.









In a follow up study, Burger and Wijnen (1999) replicated the Boers et al. study using a

larger group of participants and new stimulus words. Similar to Boers et al., the results of the

follow up study indicated that (a) the overall SRT for PWS was slower than that for PWNS, (b)

SRTs were faster in the homogeneous condition compared to the heterogeneous condition, and

(c) the priming effect was more prominent when the priming word and the target shared the

initial CV than when they only shared the initial C. However, Burger et al. also reported that they

failed to replicate the interaction effect between group, prime type, and condition reported in the

Boers et al. study. In other words, the authors reported that both groups benefited from priming

similarly and showed a similar and larger effect of priming in the CV condition compared to the

C condition.

In a more recent study, Melnick and Conture (2003) also examined the effect of priming on

phonological encoding in general and speech reaction time in particular in preschool children

who do and do not stutter. The authors hypothesized that "if CWS do, in fact have an impaired

ability to quickly phonologically encode as the CRH suggests, then their ability to benefit from

phonological priming may not be as apparent or as great as that of CWNS." (p. 1438). The study

included 18 children who stutter (CWS) and 18 children who do not stutter (CWNS) (3-5 year

old). Unlike Wijnen and Boers' s (1994) study design, Melnick and Conture' s (2003) priming

condition had three levels: No prime, related phonological prime, and non related phonological

prime. In the no prime condition, the children saw pictures of common obj ects and were asked to

name the obj ects as fast as they could. In the related priming condition, the children saw and

described the same obj ects presented in the first condition; however, 500 ms prior to picture

presentation, the children heard either a related or unrelated "auditory stimulus". The auditory

stimulus either shared (related prime) or did not share (unrelated prime) the first CV or CCV of









the picture (authors did not provide examples). Participants were instructed to name pictures as

fast as they could and SRT was recorded during the trials. Similar to the results reported by

Wijnen and Boers (1994), the results revealed that children in both groups exhibited faster SRT

during the related versus unrelated prime conditions. The results also indicated that the older

children in both groups exhibited an overall faster SRT than the younger children, and that the

older children in both groups exhibited faster SRT during the related priming condition than the

two other conditions. Post hoc analyses revealed that there was a difference between the two

groups in the relationship between children's scores on tests of articulatory mastery and their

reaction time scores. Specifically, the authors reported that the higher the scores of the children

who do not stutter on articulatory mastery tests, the faster their reaction times were during

picture naming. This relationship was not observed in CWS. Based on these results, Melnick and

Conture concluded that "preschool children who stutter, as a group, may have somewhat less

well developed articulatory systems than preschool children who do not stutter." (p.1428). Most

importantly, the authors suggested that phonological priming methodology could help in

examining the "speech-language" planning and production of young (i.e., 3-5) year-old speakers.

Lexical/Semantic Priming and PWS

Priming methodology has also been recently used to examine lexical/semantic-encoding

abilities of CWS (Hartfield & Conture, 2006; Pellowski & Conture, 2005). Pellowski and

Conture examined the effect of priming on SRT in both CWS and CWNS. The participants in the

study were 23 CWS and 23 CWNS stutter ranging in age from 3;0 (years;months) to 5; 11. The

study included three priming conditions: no prime (i.e., no auditory stimulus was presented prior

to the target picture), a related prime condition (i.e., a semantically related prime was presented

700 ms prior to the target picture; e.g., auditory prime "boat" and target picture "car"), and a non

related prime condition (i.e., a semantically unrelated prime was presented 700 ms prior to the









target picture; e.g., auditory prime "fork" and target picture "car"). The authors reported two

findings pertaining to the priming effect: (a) SRT was significantly slower in CWS compared to

CWNS across the three priming conditions, and (b) When comparing no prime condition to

semantically related condition, responses of CWS were significantly slower than the responses of

CWNS. Based on these results, Pellowski and Conture concluded that CWS may exhibit

difficulties with lexical encoding which in turn may contribute to the speech disfluencies

produced by those children.

Similar to Pellowski et al. (2005), Hartfield and Conture (2006) examined the effect of

conceptual and perceptual properties of words on lexical retrieval in CWS and CWNS. The

participants in the study were 15 3-5 year old CWS and a matched group of CWN\S. Participants

in the study named obj ects presented under four lexical priming conditions: (a) a neutral prime-

pure tone, (b) a prime word physically related to the target word (e.g., prime word "tomato" and

target word "apple"), (c) a prime word functionally related to the target word (e.g., prime word

"bite" and target word "apple"), and (d) a prime word categorically related to the prime word

(e.g., prime word "lemon" and target word "apple"). The authors reported the three following

main findings: (a) Similar to the findings ofPellowski et al., CWS were significantly slower at

initiating fluent accurate responses than CWNS, (b) Functionally related prime and target

resulted in faster responses from CWS (but not CWNS) than physically related prime and target,

and (c) there were no significant differences between CWS and CWNS in error production.

Based on the results of the study, Hartfield and Conture suggested that the difference between

the two groups in SRT during picture naming could result from "subtle differences in various

aspects of speech-language planning" (p.320). The authors further suggested that the source of









these differences might be found in "(1) the linguistic plan, (2) the motor program, and/or (3) the

transfer of information between linguistic plan and motor program" (p. 320).

Syntactic Priming and PWS

To my knowledge, Anderson and Conture (2004) are the only researchers who used the

priming paradigm to examine syntactic processing abilities in persons who stutter. As Anderson

et al. (2004) explained, their study was motivated by consistent earlier reports of relationships

between utterance length, syntactic complexity, and fluency in CWS. The authors explained that

findings of such relationships suggest that "young CWS may experience some degree of

difficulty quickly and/or efficiently formulating morphosyntactic structures" (p.553). The

authors suggested that using the syntactic priming methodology "may help us better understand

selective aspects of the temporal component of linguistic processing in young CWS, particularly

those associated with syntactic processing" (p. 555).

In detail, Anderson and Conture used the syntactic priming methodology to examine

syntactic processing in CWS. The participants in the study were 16 CWS and 16 CWNS stutter

between the ages of 3;3 and 5;5. The authors used a syntactic priming task that they described as

"an age-appropriate version of the syntactic priming paradigm [which was used by Bock and

colleagues in several psycholinguistic studies]" (p.552). The authors explained that latencies of

responses (i.e., SRT) for the participating children were measured during a syntactic priming task

that consisted of two conditions-a no prime condition and a syntactic prime condition. An

unrelated syntactic prime condition was not used in the study because a pilot study showed that

when such a condition was used with younger children, the unrelated primes seemed to affect the

probability of the children later producing the target syntactic structures and thus resulting in the

children not producing enough samples of the target sentence structure.









The experimental pictures used in the study could be described by using simple, active,

affirmative, declarative (SAAD) sentences, such as 'The girl is petting the cat.' The other/filler

pictures could be described using other types of sentences such as negatives. In the no-prime

condition, children were presented with pictures and were instructed to describe each picture as

quickly as possible. Each picture was presented for 3000 ms, and children's responses were

recorded so that SRT could be assessed for each of the responses. During the no-prime condition,

children saw the same pictures used in the syntactic prime condition however, 2000 ms prior to

the onset of picture presentation, a priming sentence that exhibited a syntactic structure similar to

that of the target sentences was auditorily presented.

Consistent with studies of typical speakers (e.g., Smith & Wheeldon, 2001), the results of

the study revealed that SRT for both groups was shorter during the related syntactic prime

condition compared to the no-prime condition (in CWS: shorter by approximately 212 ms, and in

CWNS: shorter by approximately 51 ms). Results further revealed that there were significant

differences between the two groups in the priming effect. Specifically, CWS were significantly

faster by approximately 212 ms in the syntactic-priming condition than in the no-prime condition

compared to CWNS (51 ms) who did not show a significant difference in SRT between the two

conditions. Based on these results, and among other conclusions, Anderson and Conture

concluded that CWS seemed to benefit more from syntactic priming than CWNS.

Summary

In summary, both theoretical and empirical evidence suggest that stuttering might result

from difficulties in speech formulation processes prior to speech production. In addition,

evidence from numerous studies suggests that increased linguistic loads during the speech

production tasks seem to affect the temporal and fluency characteristics in PWS differently than

in PWNS. Experimental designs that incorporate methodologies such as priming may offer a









better understanding of how increased linguistic demands affect speech formulation and

production in PWS. Methodologies such as the syntactic priming methodology and the material

used by Bock and colleagues during the study tasks could be specifically helpful in such an

investigation because this material has the following appealing features:

* The material used by Bock and colleagues in the syntactic priming studies has been used
for over two decades and has provided reliable results about syntactic processing in typical
adult speakers. This in turn is expected to provide us with a basis for comparing the results
we obtain in the present study with both PWS and PWNS to the results of earlier studies
that used this methodology.

* A series of follow-up studies by Bock and colleagues clearly established that the carefully
prepared material which was used in the syntactic priming studies insured that the syntactic
priming effect observed in studies using that material is independent of lexical (Bock,
1986) and thematic/semantic effects (Bock & Leobell, 1990).

* A more recent study with CWS (Anderson and Conture, 2004) showed that syntactic
priming is a feasible procedure for studying speech-language planning and production in
PWS.

* The sentences used by Bock and colleagues in the syntactic priming studies include
sentences that cover a variety of syntactic structures (e.g., transitive, dative, two-clause
embedded and conjoined) and syntactic complexities (i.e., transitive-low, dative-
intermediate, and two-clause- high syntactic complexity) and thus can be used to examine
the effect of different sentence types on speech production processes in both PWS and
PWNS (a detailed review of those sentence structures is presented in Chapter II).

Thus, it seems reasonable to suggest that using the syntactic priming methodology,

particularly the materials used by Bock and colleagues in their syntactic priming studies, could

provide important information and initial insights into the temporal and syntactic production

abilities of persons who stutter and in turn yield useful information about the factors that are

associated with disfluent speech.

Research Questions

The central question for this study is as follows: Is there a significant difference between

adults who stutter and adults who do not stutter in syntactic formulation and production under

conditions of increased linguistic demands? The four following questions are related to this









broad issue and are specifically related to the effect of syntactic structure and syntactic priming

on speech fluency and speech reaction time in responses produced by PWS and PWNS.

Question One

Does syntactic structure affect speech fluency of persons who stutter and persons who do

not stutter? If so, is speech fluency of persons who stutter affected differently by syntactic

structure than that of persons who do not stutter?

Question Two

Does syntactic structure affect speech reaction time in persons who stutter and persons

who do not stutter? If so, are speech reaction times of persons who stutter affected differently by

syntactic structure than those of persons who do not stutter?

Question Three

Does syntactic priming affect speech fluency in persons who stutter and persons who do

not stutter? If so, is the percentage of fluent responses produced by persons who stutter affected

differently by syntactic priming than that of persons who do not stutter?

Question Four

Does syntactic priming affect speech reaction time in persons who stutter and persons who

do not stutter? If so, are speech reaction times of persons who stutter affected differently by

syntactic priming than those of persons who do not stutter?









CHAPTER 2
IVETHOD S

Participants

15 PWS and 15 PWNS participated in the study. The mean age for the participants who do

not stutter was 32.2 years (S.D. = 12.76 years; range 21 59) and the mean age for the

participants who stutter was 32.4 years (S.D. = 14.05 years; range 19 59). Persons in the two

groups were matched for years in school (+/- 3 years), age (+/- 2 years) and gender (10 male, and

4 female participants in each group). Overall, 14 out of the resulting participant pairs were also

matched for age and years in school. One pair was not matched for years in school nor age

(PWS: 12 years in school, 52 years old; PWNS: 16 years in school, 24 years old). Several

mechanisms were used to recruit participants. These included contacting patients on the waiting

list for diagnostic evaluations at the University of Florida Speech and Hearing Clinic and at

several speech and hearing clinics in the State of Florida, posting signs on the University of

Florida campus, and by making announcements in university classes, and posting advertisements

in local and college newspapers.

Prior to participating in the study, participants completed several formal and informal tasks

that were used to determine their eligibility for the study. These tasks are explained in detail

below. Participant characteristics are presented and summarized in Table 1.

Inclusion and Exclusion Criteria

Participant inclusion and exclusion criteria were as follows:

General inclusion criteria: To be included in the study, participants had to:

* be native speakers of American English.

* have a negative history of any medical, neurological, or emotional conditions that might
influence their performance during the study.










* have no active speech or language impairments other than stuttering and have no past
history of language-related special services at any point during their academic careers.

There were 16 participants who stutter and 22 participants who do not stutter who initially

participated in the study. From this initial group, 3 PWS and 8 PWNS were excluded from the

analysis for the following reasons:

* One participant who stutters completed the protocol for the study; however, the
participant' s data were excluded from final analysis because he failed to meet the inclusion
criteria described above (i.e., had a positive history of neurological problems).

* Although 22 participants who do not stutter completed the study protocol, data for only 15
participants in this group was used in the analyses. These participants were selected to
insure a good age, education, and gender match with the participants who do not stutter.
The excluded participants either did not match the PWS in age or in education. In addition,
several of the participants in the PWNS group had an academic background in linguistics
and data from meetings with those participants were not included from the analyses. This
was done to eliminate any bias in the results based on those participants' prior knowledge
or exposure to procedures and data similar to the ones used in the present study.

Material

Measurements of speech reaction times, speech fluency, and sentence structure were made

from audio and video recordings, which were collected during a syntactic priming task that

lasted approximately 50 minutes. Details about the study material (i.e., pre-study tasks and main

task), data preparation, and study procedures are described immediately below and are also

summarized in Table 1.1.

Pre-Study Material Used with all Participants

Background survey (Appendix A)

All participants finished a survey aimed at getting background information about their

speech and language abilities. Information included the general speech and communication skills

and educational background. There were two forms for this survey: one used with the PWS and

the other with the control group (i.e., PWNS). The survey asked participants about any speech,

language, vision, or hearing problems they might exhibit. The survey also asked the participants









about their education level, type of school they went to (i.e., two year vs. four-year College) in

addition to the highest educational degree earned by the participant.

WAIS vocabulary

The Vocabulary subset of the WAIS test is usually used as a general measure of

intelligence and of verbal comprehension. It can also provide information about the participant' s

education and life experiences (see Zimmerman and Woo-Sam, 1973 for a detailed description of

this test). The test consists of 35 words, which the participant is asked to define. The words on

the test are listed in order of difficulty and have equal numbers of nouns and verbs, plus several

adj ectives and one adverb. As explained by Zimmerman and Woo-Sam (1973), "Vocabulary

indicates sensitivity to new information and ideas and the ability to store and associatively

regroup these as the occasion demands."(p. 108).

The mean score for the PWNS was 62.79 out of 70 and the mean score for the PWS was 54

out of 70. The results of the t tests for the WAIS vocabulary test indicated a significance

difference between the two groups (t (26) = 2.538, two-tailed p = 0.017).

Digits forward and digits backward

The tests measure immediate auditory recall, attention, concentration, and the ability of the

participant not to be affected by distracters. In general, the point of interest in the tests is to

identify whether the participant' s responses on the test show any discrepancy between digit

forward and digit backward recall and if that is the case, the direction of this discrepancy (i.e., in

favor of digit forward or digit backward).

The mean score on the Digits Forward test for the group was 10.93 out of 14 and the mean

score for the PWS was 9.50 out of 14. The results of the t tests indicated no significant difference

between the two study groups on the digits forward test (t (26) = 1.286, two-tailed p = 0.210).









The mean score on the Digits Backward test for the PWNS was 9.070ut of 14 and the mean

score for the PWS was 7.29 out of 14. The results of the t tests indicated no significant difference

between the two study groups (t (26) = 1.632, two-tailed p = 0. 115).

Digit ordering

This test is used to assess linguistic working memory and/or language processing skills (see

Maryellen C. MacDonald, Amit Almor, Victor W. Henderson, Daniel Kempler, and Elaine S.

Andersen, 2001 for details).

The mean score for the PWNS group was 20.43 out of 24 and the mean score for the PWS

group was 17.21 out of 24. The results of the t tests indicted a significant difference between the

two study groups on the digit ordering test (t (26) = 2.5380, two-tailed p =.010).

Pre-Study Material Used with PWS

Stuttering severity level task (Appendix A)

Two speech samples, each of which was approximately 300 syllables long, were elicited

from each of the participants in the experimental group. This task was used to provide additional

information about the speakers' overall level of speech fluency. To elicit the speech samples,

each of the PWS was asked to talk for 3 minutes about the plot of a movie they had seen

recently, and then for another three minutes about a pleasant experience they had. The resulting

speech samples were audio and video recorded and later analyzed for (a) types of disfluency

produced (i.e., 'repetitions and prolongations' such as the prolongation of sounds, or the

repetition of sounds or parts of words, the repetition of monosyllabic words, multi-syllabic

words, or phrases, or 'other' disfluencies such as revisions of words or phrases and interj sections)

(b) percentage of syllables stuttered in a speech sample of 100 syllables.









Self-rating scale (Appendix A)

In addition to the above pre-study tasks, each of the participants who stutter finished a self

rating scale intended to get general information about each participant' s stuttering severity

during everyday situations.

Preparation and Description of the Material Used in Syntactic Priming Task

The sentences and pictures used in the present study were originally developed and used in

syntactic priming studies conducted by Bock and colleagues and later adapted and used by Dr.

Lori Altmann at the Language over the Lifespan Laboratory at the University of Florida (Bock,

1986, 1989; Bock & Griffin, 2000; Bock, Leobell, & Morey, 1992). The material consisted of

priming sentences and associated picture representations, plus a variety of filler sentences and

their associated picture representations. The remainder of this section describes those sentences

and their associated pictures in detail. It is worth mentioning that although I provide a detailed

description of the study material below, the material used was generously shared by Dr. Lori

Altmann at the University of Florida who used the same sentences and pictures to elicit

responses in studies using methodology similar to the one used in the present study.

The study material consisted of 54 sets of priming sentences, which were paired with 54

pictures of events. A given priming sentence set was classified as transitive, dative or two-clause

based on (a) the main verb type in each sentence, (b) the syntactic complexity of each sentence

as described below in detail (See Appendix B for a list of all the priming sentence sets). The

following section provides a detailed description of each of these three sentence types.

Transitive priming sentences and pictures

There were eighteen sets of transitive priming sentences each of which included an active

sentence, a corresponding full passive (a 'by' passive) sentence, and a locative sentence. The

locative sentences were used in Bock and colleagues' studies as a control condition to prime the









passive sentences since the two sentence types had a similar phrase structure (see below). The

structure of the active sentences in the transitive priming set was as follows: a subj ect noun

phrase, followed by a transitive verb and a direct-obj ect noun phrase as in Example (1):

(1) The construction worker drove the bulldozer.

The structure of the corresponding passive sentence in the transitive priming set was as

follows: subj ect noun phrase, followed by a full passive verb, and the preposition by followed by

a prepositional noun phrase as in Example (2)

(2) The bulldozer was driven by the construction worker

The structure of the locative sentence in the transitive priming set was as follows: a subj ect

noun phrase, followed by a verb in the present participle form, and the preposition by followed

by a prepositional noun phrase as in Example (3):

(3) The construction worker was digging by the bulldozer.

Each of the transitive priming sets was matched with one target picture intended to elicit

transitive verbs (active and passive). Each of the transitive verb pictures depicted an action

involving an agent and an obj ect undergoing the action (i.e., each transitive picture included two

actors in the pictured event). The actions depicted in the pictures included driving, destroying,

striking, chasing, kicking, towing, stinging, hitting, kissing, pushing, and squirting (e.g.,

lightning striking a church, and a wrecking ball destroying a building). The action in each of the

transitive pictures could be described using either an active transitive or a passive transitive

sentence structure. To illustrate, one of the transitive sets included the following three sentences:

Active: 'The lumberj ack struck the giant redwood tree'; Passive: 'The lumberj ack was struck by

the giant redwood tree'; Locative: 'The lumberjack was eating by the giant redwood tree.' The










picture matched with this set was of a fireman rescuing a baby from fire (See Appendix B for a

full list of the transitive sentence sets and target picture that was matched with each set).

Dative priming sentences and pictures

There were eighteen sets of dative verb sentences each of which included a prepositional

dative sentence, a corresponding double-obj ect dative sentence, and an obj ect complement

sentence. The obj ect complement sentences were used in Bock and colleagues' studies as a

control condition to prime the double-object sentences since the two sentence types have a

similar phrase structure as described below. The structure of the prepositional dative sentence

was as follows: a subj ect noun phrase, followed by a dative verb, which was followed by a

direct-obj ect noun phrase, and a prepositional noun phrase beginning with to, as in Example (4):

(4) The children sang a song to their babysitter.

The structure of the corresponding double-obj ect sentence in the dative priming set was of

a subj ect noun phrase, followed by a dative verb, which was followed by a direct-obj ect noun

phrase, and an indirect-object noun phrase, as in Example (5):

(5) The choir sang the wedding guests a new hymn.

The structure of the control obj ect complement sentence in the dative priming set was as follows:

a subj ect noun phrase, followed by a dative verb, which was followed by a direct-obj ect noun

phrase, and an obj ect complement as in Example (6):

(6) The choir considered the new hymn their favorite song.

Each of the sentences in the dative priming sets was matched with a target picture intended

to elicit dative verbs. Each of the dative verb pictures depicted an action involving an agent, an

obj ect undergoing the action, and a recipient of the action (i.e., each dative picture included three

actors in the pictured event). The actions depicted in the pictures included giving, showing,

throwing, passing, reading, serving, and handing (e.g., a librarian giving a book to a boy, and a









waitress serving drinks to a man). Each of the dative priming pictures could be described using

either a double-obj ect dative or a prepositional-dative sentence structure. To illustrate, one of the

dative sets included the three following sentences: prepositional dative 'The diplomat took the

secret documents to the president', double-obj ect dative 'The diplomat took the president the

secret documents,' and obj ect complement 'The diplomat made delivering the secret documents

his primary mission.' The picture matched with this set was of a girl and a boy showing a picture

to a teacher (See Appendix B for a full list of the dative sentence sets and target picture that was

matched with each set).

Two-clause priming sentences and pictures

There were eighteen sets of two-clause sentences each of which included a conjoined

sentence (i.e., two independent clauses joined using the coordinating conjunction 'and', or the

subordinating conjunction 'because'), a center-embedded (subject-subject) relative clause, and a

right-embedded (object-obj ect) relative clause (i.e., one independent clause and one center or

right embedded relative clause) as in Examples (7), (8), (9) respectively:

(7) The duck is kicking the girl and the girl is kicking the boy.

(8) The girl that is kicked by the duck is kicking the boy.

(9) The duck is kicking the girl that is kicking the boy.

Each sentence in the two-clause priming sentence sets was coupled with a picture that was

intended to elicit sentences that can be described using either a conj oined clause, a center-

embedded clause, or a right-embedded clause. To illustrate, one of the two-clause sets included

the three following sentences: conj oined 'The girl smelled the flower and it reminded her of her

grandmother,' center-embedded 'The girl that smelled the flower was very pretty,' and right-

embedded 'The girl smelled the flower that was very pretty.' The picture matched with this set









was of a man pulling a woman who was pulling a dog (See Appendix B for a full list of the

dative sentence sets and target picture that was matched with each set).

Filler sentences and pictures

In addition to the priming sentences and target pictures described above, there were 127

filler sentences and pictures (See Appendix B for a list of the filler sentences). The filler pictures

were similar in style and preparation to the experimental pictures but depicted actions that are

commonly described with sentence forms different from those used in the priming sentences.

Forms used in the filler sentences included reflexives, and predicate adj ective sentences as in

Examples (10) and (11) respectively:

(10) The surgeon cut himself with a scalpel

(11) The books were expensive.

Using the above described sentences and pictures, three lists were developed and used in

the study. Each of the lists contained 252 priming trials that consisted of random presentations of

a priming sentence, priming picture, filler sentence, or filler picture. Each of the priming

sentences was immediately followed by a priming picture. Thus a typical sequence of stimuli in a

priming trial was: filler sentence, filler picture, priming sentence, priming picture. The filler

sentences were randomly interspersed within each list and could either have the sequence of one,

two, or three consecutive filler sentences followed by one, two, or three filler pictures. There

were ten filler sentences between each two priming sentences.

The transitive, dative, conj oined, and embedded priming sentence sets alternated in each

list so that an equal number of each sentence structure occurred in each list (6 sentences each of

active transitive, passive transitive, locative, double-object dative, prepositional dative, object

compliment, conjoined, center-embedded, and right-embedded, resulting in 9 sentence types x 6

instances of each type, which equaled a total of 54 priming sentences and 54 corresponding










target pictures in each study list). The different sentence structures alternated in the lists in a way

that ensured participants did not encounter two priming sentences or two target pictures of the

same type in successive priming trials. Each list included only one sentence type from each set.

For example, list A included the transitive active sentence 'The lumberj ack struck the giant

redwood tree,' list B included the corresponding passive sentence from that set 'The lumberjack

was struck by the giant redwood tree' and list C included the locative sentence of that set 'The

lumberj ack was sitting by the giant redwood tree.' The same procedure was followed for the

distribution of all other sentence sets among the three study lists. The order of the priming trials

was the same in the three lists.

Following the data collection session, each participant' s audio and video-recorded speech

sample was transcribed orthographically and then coded to get specific information for each

sentence. Information about sentence type, speech reaction time, and fluency for sentences

produced by each of the participants was determined as described below under Data Analysis.

Apparatus

A digital voice recorder (Olympus WS-100) was used to record responses during the pre

and main study tasks. During the experiment, the stimuli were presented on a laptop computer

(Dell Inspiron 15150) and the speech productions were recorded using a lightweight headset

microphone (Optimus 33-3012) that was approximately two inches from the speaker's mouth.

The study material was presented using Direct-RT (DirectRT 2006 2.16) a graphical experiment

generator which is capable of managing and analyzing data during time controlled experiments

such as the one used in the present investigation. Direct-RT also recorded in milliseconds each

participant' s SRT for picture description during the study task.









Procedures


Data Collection

Each of the percipients was seen individually for one session divided into two parts. Upon

successful completion of eligibility testing in the first part of the session (described above under

Material), participants finished the second part of the session (i.e., the syntactic priming task)

during which SRT and fluency data collection took place. The first part of the session took 20 to

25 minutes to complete, and the second part took 45 to 60 minutes to complete. It is worth

mentioning that the participants finished a second task (i.e., sentence generation task); however,

data and results for responses during that task are not reported in this paper. The following is a

detailed description of the procedures followed during the syntactic priming task.

I described the task to the participants and told them that the first 9 items were practice

material. The participant saw either a picture on the computer screen or the sentence, "Listen and

Repeat." When they saw a picture, they were instructed to 'make up' a one-sentence description

of the event pictured and say it aloud as soon as possible after the picture appeared. They were

also instructed not to use pronouns in their picture. This instruction was added to 'motivate' the

participants to produce the appropriate number of noun phrases representing the main actors in

the depicted events such as 'The bear is kicking the girl and the girl is kicking the boy' instead of

a partial description such as 'They are kicking each other' when a pronoun is used in that

response. When the participants saw the sentence "Listen and Repeat," I read a sentence and the

participants were instructed to repeat the sentence I just read. The participants were told that the

experimenter could read the sentence again if it was not clear the first time.

If the participant responded successfully to the first 9 items on the list, the experimental

sentences and pictures were presented. In instances were participants' response to a trial item

was not accurate or when the participant seemed confused about the nature of the expected










response, I explained the error made by the participant on that trial and asked the participant to

repeat his/her response to that trial following the correct instructions. After finishing the first 9

items in the task, the main task items were presented individually on the computer screen.

For each trial, the instruction "Listen and Repeat" and the picture remained on the screen

until the participant repeated the sentence or finished the description of the pictured event.

Immediately after each response, the participants were asked if they've heard the sentences or

seen the pictures before within the course of the experiment by answering the questions "Have

you heard this sentence before?" that immediately followed their sentence repetition and the

question "Have you seen this picture before?" which immediately followed their picture

description. Participants answered the question by responding "yes" or "no." The questions were

used as part of a "recognition memory test" following Bock (1986) procedures to "minimize

subjects' attention to their speech and its structural features (p. 360). The question remained on

the screen until the participant responded to that question. I then pushed a mouse button which in

turn triggered the software program to move to the next trial.

The participants were told when the practice items were presented that the task is rather

long (on average took participants about 50 minutes to finish) and that they would be offered

several opportunities to rest, if necessary, during the task. Participants were also told that they

could ask for a break at any time during the task.

Data Preparation for Analyses

To summarize the method used to generate the data for this study, I transcribed verbatim

all sentences produced by the participants. Each sentence was then coded as a target sentence or

filler sentence. All filler sentences were excluded from further analyses. The remaining sentences

were further coded for general descriptive characteristics (i.e., number of syllables and words in

each sentence), grammaticality (i.e., whether the sentence was grammatical or not), sentence










type (whether the sentence was transitive, dative, two-clause, or other), and for fluency measures

(i.e., whether the sentence (a) was all fluent, (b) contained a speech disfluency in any part of the

sentence, (c) contained a speech disfluency in the first noun phrase in the sentence (d) contained

a speech disfluency in the first word of the first noun phrase in the sentence; as well as the

number of disfluencies in each sentence, and the type of disfluencies in each sentence).

The transcripts were used to determine response fluency (i.e., fluent or disfluent;

'repetitions and prolongations,' or 'other' types of disfluencies, and the location of disfluency

within an utterance) and response structure (i.e., Transitive: active, passive; Dative: prepositional

dative, double-object dative; Two-clause: conjoined, embedded). The following are the details of

data preparation for analysis.

Data excluded from the final analyses

Three main data analyses were of interest in the present study: fluency analysis, sentence

type analysis, and SRT analysis. For each analysis, several responses were excluded from the

final analyses based on criteria following either procedures used in other studies (e.g., Bock et al.

1986, Logan, 2003). Out of 1620 possible responses for the two groups (i.e., 54 responses x 30

participants), 1617 sentences were actually produced and a total of 328 (20.28%) sentences were

excluded from the final analyses based on the application of a number of exclusion criteria which

are described below.

The first of the sentence exclusion criteria concerned elicitation and response problems

(i.e., problematic responses). A 'problematic' response was defined as one that was associated

with (a) an unrelated vocal response preceding the onset of the target verbal response (coded as a

'false start' e.g., a yawn, cough, or laugh by the participant, and responses that started with a

question or a comment by the participant that was not part of the target response), (b) program

error (i.e., the software program used to present the priming pictures skipped an item), (c)










presentation of the wrong priming sentence by the experimenter (i.e., I read the wrong priming

sentence or one that was not associated in its syntactic structure to the target picture). The total

number of problematic responses was 107 (6.61 %) responses (64 false starts, 32 program skips,

and 1 1 wrong sentences). Thus, following application of the first of the sentence exclusion

criteria, 1510 sentences remained for analysis.

The second of the sentence exclusion criteria concerned response grammaticality. An

'ungrammatical response' was defined as one that was missing a main verb or parts or a verb

(e.g., "A boy running,") or exhibited inaccurate word form or word choice (e.g., "The bulldozer

is demoralizing the building," "The pedestrian got hit by a rescue,", and "The turtle wet the cat").

The total number of ungrammatical responses was 1 12 sentences (6.92 % of 15 11 remaining

responses). Thus, following application of the second of the exclusion criteria, 1398 sentences

remained for analysis.

The third of the sentence exclusion criteria concerned examination of speech reaction time

for each sentence produced by the participants. Recall that SRT in milliseconds for each

participant was captured by the computer and was defined as the time from the onset of the

presentation of the priming picture on the computer screen to the onset of verbal response by the

participant captured by the voice activated microphone. To eliminate the effect of outlier SRT

values, and after consultations with two committee members, I trimmed the SRT data by

including only the SRT values that fell between 500 milliseconds, at the low end, and three

standard deviations above the SRT mean value for a given participant, at the high end. Thus, all

SRT values for grammatical sentences below 500 milliseconds as well as SRT values that fell 3

standard deviations above the mean SRT value for a participant were excluded from subsequent

analyses. The total number of excluded SRT outliers was 63 sentences (3.89 % of 1398









remaining responses). Following the application of the third exclusion criteria 133 5 sentences

remained for analysis.

The fourth and last of the exclusion criteria concerned overall participant performance.

One participant in the experimental group produced only two grammatical sentences out of the

54 possible responses thus resulting in only 2 usable sentences. As a result, responses for that

participant were excluded from the Einal statistical analyses, as were response for the matched

participant from the control group. Thus, in all, a total of 46 possible responses (produced by

those two participants) or 2.84 % out of 1335 remaining responses were excluded from

consideration. Accordingly, statistical analyses described in the subsequent sections are based

upon sentences produced by the remaining 28 participants (14 PWNS and 14 PWS).

After all the sentence exclusion criteria were applied, the data file used to conduct all

subsequent analyses included only grammatical sentences, that were (a) entirely accurate (i.e.,

they contained no false starts or evidence of program or experimenter errors) and (b) free from

outlying SRT scores. Thus, 1289 responses were included in the Einal analyses. This number

represented 79.71 % of the total set of responses that the participants originally produced.

General descriptive analyses

Each participant had the opportunity to produce 252 responses: 108 critical trials (i.e.

repetitions of 54 priming sentences and generation of 54 sentences to describe target pictures),

and 144 Eillers (i.e., repetitions of 72 Eiller sentences and generation of sentences to describe each

of the 72 Eiller pictures). Only picture descriptions during the critical trials were of interest in this

study, thus the sentence fluency analyses were based on the 54 possible sentences generated by

each participant to describe the priming pictures. All other responses (i.e., repetitions of priming

and filler sentences and descriptions of filler pictures) were excluded from the Einal analyses.










Data preparation for speech fluency analyses


Each sentence produced by the participants was coded as to whether it was fluent or

disfluent. Fluent sentences were those that did not contain 'repetitions and prolongations' as

defined in (1) to (7) below and 'other' disfluency types as defined in (8) and (9) below:

1 Sound or syllable repetition: Instances at which the participant repeated a sound or
syllable in a given word as in "The b boy is giving the girl a valentine's card," and "The
bear is hiding behind the tree while the boy is look looking at the duck."

2 Monosyllabic word repetition: Instances at which the participant repeated a one-syllable
word as in "The the man is cleaning the window".

3 Polysyllabic word repetition: Instances at which the participant repeated a word that
contained more than one syllable as in "The hostess is serving serving drinks to her
guests".

4 Phrase repetition: Instances at which the participant repeated a string of words within a
sentence as in "The bee the bee is stinging a man".

5 Inaudible sound prolongation: Instances at which the participant seemed to start a word
and "block" or exhibit difficulty moving from the initial sound or part of that word to the
next sound as in "The p policeman is handing the man a ticket".

6 Audible sound prolongations: Instances at which the participant started a sound and
vocalization of the sound could be heard as in "The fff firefighter is rescuing a baby from
the fire".

7 Broken words: Instances at which participants produced a sound or syllable repetition or
an inaudible sound prolongation at syllable divisions within a word as in The
construction worker drove the bull do dozer," and The boy is hold ding a girl's hand".

8 Other" disfluency types: Interj sections and revisions as in "Um, The construction worker
drove the uh bulldozer".

9 Revisions: Instances at which the participant responded with a phrase or a sentence,
stopped, and changed their choice of that phrase or sentence as in A clier, a fireman is
climbing out of a building" and "A girl is squirting a boy with a gun, with a squirt gun",
and "A man and a woman, oh no, A man is looking at a baby and a woman is looking at
the man".

Disfluent sentences produced by each group were further coded in terms of the position of

the disfluency (e.g., 1st word, 1st noun phrase, etc.) within the response. This analysis was

performed to be used in providing general descriptive data for fluency and was also used to guide









the manner in which statistical analyses for the SRT analyses were to be conducted (described in

more detail below). Sentences that were produced with no disfluencies were coded as "fluent".

Overall, participants produced 646 entirely fluent sentences (50. 11 %) and 643 disfluent

sentences (49.88 %). In terms of group descriptive statistics, 76.22 % of sentences produced by

PWNS were entirely fluent and 23.92% were disfluent. As expected, PWS showed a pattern that

was nearly opposite the one that was observed for PWNS, that is, only 21.59% of the sentences

produced by PWS were entirely fluent and 78.40 % were disfluent.

Data preparation for sentence type analyses

Following data preparation procedures similar to those used for the fluency analyses,

sentence type analyses was based on the same 54 sentences generated by each participant to

describe the priming pictures. All other responses (i.e., repetitions of priming and filler sentences

and descriptions of filler pictures) were excluded from further analysis. Each of the 54 sentences

produced by a participant was assigned to one of the following categories: 'Transitive' as in

examples (1) and (2) above under 'Transitive priming sentences and pictures', 'Dative' as in

examples (4) and (5) above under 'Dative priming sentences and pictures,' and 'Two-clause' as

in examples (7), (8), and (9) above under 'Two-clause priming sentences and pictures.' If a

sentence was coded as 'Transitive,' it was further coded as either 'active' or 'passive.' To be

coded as an active, the sentence had to have an acceptable corresponding passive sentence form.

A passive form had to include a passive verb (i.e., a main verb proceeded by a form 'be' or 'get'

and followed by a 'by-phrase' (i.e., prepositional phrase starting with the preposition 'by'). For

example, an active sentence was "The dog is chasing a mailman" and the corresponding passive

sentence was "The mailman is chased by the dog" (see 'Transitive priming sentences and

pictures' above for detailed description and examples of the active and passive sentence types).









Passive sentences that did not include the by-phrase were coded as 'other' (e.g., "A man is bitten

in the arm").

If a sentence was coded as 'Dative', it was further coded as either 'prepositional dative', or

'double-obj ect dative'. To be coded as 'Dative', the sentence had to have an acceptable

corresponding double-object sentence form (see 'Dative priming sentences and pictures' above

for detailed description and examples of the 'prepositional' and 'double-object' sentence types).

Dative sentences had to include appropriate numbers of noun phrases indicating the three main

actors in the depicted action (e.g., "The girl (1 is giving the man (2) a paint brush (3)"). Dative

sentences that did not include all three main actors (e.g., "The girl ") read to the boy(2) a

compared to The girl "), read a story(2), to the boy(3)") were coded as 'other'. An additional

category: 'dative + two-clause' was added to the 'Dative' sentence type analysis. The 'dative +

two-clause' category was added to show that in some instances, the 'Dative' and 'Two-clause'

categories were not mutually exclusive. If a sentence was coded as 'two-clause', it was further

coded as either 'conjoined' or 'embedded.' (see 'Two-clause priming sentences and pictures'

above for detailed description and examples of the 'conjoined' and 'embedded' sentence types).

Finally, other sentences structures (i.e., non-transitive, reflexives, and predicative adjective

sentences as in examples (10), (11), and (12) respectively above under Filler sentences pictures

were coded as 'other'. The number of sentences produced per each examined sentence type is

illustrated in Figure 2-1.

Pertaining to response length for each of the examined sentence types and as illustrated in

Figure 2-2., on average, and for participants in the two study groups, transitive sentences were

shortest in terms of both words (PWNS = 7.97, PWS = 7.74) and syllables (PWNS = 10.26, PWS









= 9.96) number, and two-clause sentences were longest (words: PWNS = 12.60, 12.63; syllables:

PWNS = 15.43, PWS = 1620).

Data preparation for SRT analyses

SRTs for each of the sentences were coded online by the Direct-RT program. SRT was

measured (in milliseconds) and defined as the time of picture (target or filler) presentation to the

time of acoustic onset for participant' s response. Recall that during picture description trials for

both the target and filler pictures, the Direct-RT program generated and controlled the time at

which pictures were presented. The program also recorded the latencies of the participant' s vocal

response, in milliseconds, for each of the pictures.

The overall mean SRT for PWNS was 2667.20 ms (S.D. 1252.28 ms) and for the PWS,

was 2596.78 ms (S.D. 1532.33 ms). Based on these values, SRT for PWS was on average 70.42

ms shorter than SRT for PWNS.

In terms of data preparation for the SRT analyses per sentence type, it is worth mentioning

that deciding on which responses to include in the SRT analyses was not entirely obvious. The

difficulty in making this decision lies in the fact that when earlier studies used sentence structure

priming methodology to examine structure priming effects in addition to SRT data, almost all

these studies used a very rigid definition of what a fluent response is. For example, in Bock' s

studies (e.g., Bock, 1986; Bock 1989) that I reviewed, whenever a reference was made to what a

fluent response is, the definition stated that such a response is one that was produced with no

entirely fluent (i.e., exhibited no speech disfluencies such as interj sections, revisions, hesitations,

or sound or syllable repetitions). Note however, that none of the above-referenced studies by

Bock examined SRT and fluency in PWS. As would be expected based on previous research,

PWS on average produce significantly more disfluent responses than PWNS. As a result, when

comparing sentence production between these two groups, and when participants in the two










groups are given equal opportunities to produce a similar number of responses, a percentage of

those responses will be disfluent; however, a substantial number of the disfluent responses will

be produced by the stuttering group. If such disfluent responses are excluded from a given

analysis, this will in the end result in the loss of several data points for participants in both

groups but substantially more for the participants in the stuttering group. In this study, out of 673

responses produced by PWNS, 513 (76.52%) were entirely fluent compared to 616 responses

produced by PWS out of which only 133 (20.25%) were entirely fluent. Clearly, if this rigid

definition of fluency is used in the present study, it would be necessary to exclude at least 4 of

the 14 PWS and the sample means for many other participants would be based on fewer than 10

data points. For example, looking at the range for how many fluent transitive sentences were

produced by each group, PWNS on average produced 13 fluent transitive sentences (range 6-18).

PWS on the other hand produced on average about 4 fluent transitive sentences (range 0-10). The

averages for the number of fluent dative and two-clause sentences show a similar pattern.

Compare this to when a more "lenient" definition of a fluent response is used. In this case,

a fluent response is defined as one that exhibits no disfluencies within the first noun phrase in a

sentence (NPlF). As illustrated in Figure 2-3, when using this definition, the number of

responses that can be included in the analyses increases for the two groups, and almost doubles

for PWS (547 responses when the sentences exhibiting a fluent first noun phrase were included

in the analysis compared to only 133 when only entirely fluent responses were included). Using

the number of transitive sentences again as an example, and when looking at the range for how

many transitive sentences exhibiting a fluent first noun phrase were produced by each group,

although the increase is not that dramatic for PWNS who on average produced 13 transitive

sentences with a fluent first noun phrase, the percentage of sentences that can be included almost









doubles (40.93%) for PWS compared to when only entirely fluent sentences were examined

(20.25%).

When yet, a broader definition is used for sentence inclusion (i.e., using all sentences that

exhibited a fluent first word in the first noun phrase of a sentence), the percentage of sentences

that can be included for PWS yet increases markedly (57.3 8%; on average, 9 sentences per

PWS).

Thus, based on the above, and to obj ectively represent the results of any effects or

interactions among the different variables in the present study, in addition to including as many

responses as possible in the analyses, the reported SRT analyses and results are for sentences in

which the first noun phrase in the sentence was fluent (NPlF). It was felt that such an approach

would offer a reasonable compromise between being forced to use mean SRT values that were

based on relatively few data points and using mean SRT values that might be conflated with

sentence factors having to do with stuttering within the subj ect constituent of the first NP.

Intrajudge and Interjudge Measurement Reliability

All fluency and sentence type measures in this study were made by the author. Inter and

intrajudge measurement reliability for sentence type were conducted by randomly selecting two

participants and reanalyzing all the sentences produced by those two participants (total 108

sentences or 7 % of the data). I first recorded the 108 sentences for sentence type coding and then

a graduate student who had a background in linguistics and who was familiar with the sentence

type coding processes used in the study independently recorded each sentence for sentence type.

Intrajudge analysis showed 99% agreement (disagreement was a coding error for one sentence),

and interjudge analysis showed 98% agreement (disagreement was a coding error for two

sentences). The interjudge points of disagreement were discussed and resolved.









Inter and intrajudge measurement reliability for fluency were conducted by randomly

selecting two participants and reanalyzing all the sentences produced by those two participants

(total 108 sentences or 7 % of the data). I first recorded the 108 responses for fluency (i.e.,

'fluent' or 'disfluent') and then recorded the disfluent responses for type of disfluency (i.e.,

'repetitions and prolongations,' 'other') and position of disfluencies within response (i.e., 1 st

word, 1st noun phrase, etc.). Then an undergraduate student who was familiar with the fluency

coding processes used in the study independently recorded each response for fluency. Intrajudge

analysis showed 98% agreement, and interjudge analysis showed 94% agreement. The points of

disagreement were discussed and resolved.











Number of Responses per Sentence Type


O PWNS
H PWS


Transitive Dative Two- Other
Clause
Sentence Type


Figure 2-1. The mean and standard deviation for the number of responses produced by persons
who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) across the
examined sentence types (Transitive, Dative, and Two-Clause) in addition to
responses that exhibited syntactic structures other than the examined types (Other).


Response Length in Words


O PWNS
SPWS


Transitive Dative Two-Clause


Response Type
Figure 2-2 Mean and standard deviation for number of words produced by persons who do not
stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) for (a) Transitive
responses in the Transitive experiment, (b) Dative responses in the Dative
experiment, and (c) Two-Clause responses in the Two-Clause experiment.










Percentage of Usable Responses Based on Fluency Definition


o"100
80
0 : 60 O PWNS

So 8 40 W

20

Entirely Fluent NP1 Fluent Wordl Fluent

Group
Figure 2-3. The mean and standard deviation for the percentage of sentences produced by
persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14)
when the sentence was entirely fluent, when the first noun phrase in the sentence was
fluent, and when the first word in the first noun phrase in the sentence was fluent













Stuttering Severity Stuttering
Level Treatment
Severe No
Moderate-Severe Yes
Moderate Yes
Severe Yes
Mild Yes
Severe Yes
Moderate-Severe Yes
Moderate Yes
Moderate-Severe Yes
severe Yes
Moderate Yes
Moderate Yes
Moderate Yes
Severe Yes
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA


Table 2-1. Participant Demographics and Performance on the Pre-study Language and Memory Tests
WAIS-


ID
Number
SO2
SO3
SO4
SO5
SO6
SO7
SO8
SO9
S10
S11
S12
S13
S14
S15
NSO1
NSO2
NSO3
NSO4
NSO5
NSO6
NSO7
NSO8
NSO9
NS10
NS11
NS12
NS13
NS14


Years of Digits Forward Digits Backward Digit Ordering Vocabulary
Gender Education Max =14 Max = 14 Max = 24 Max = 70
F 14 5 5 17 54
F 15 7 5 12 62
F 16 10 7 14 60
M 17 12 7 21 50
M 12 14 14 22 61
M 13 12 6 20 56
M 17 7 6 20 60
M 17 11 12 17 66
M 17 10 9 21 68
M 12 13 4 16 40
M 13 5 8 17 38
M 17 7 6 12 34
F 19 6 4 15 43
M 19 14 9 17 64
M 17 10 6 23 58
M 17 8 6 18 58
F 20 14 13 21 69
F 17 8 8 21 48
M 21 12 13 24 62
M 16 11 8 22 65
M 17 8 8 18 62
M 20 13 12 22 69
M 18 14 8 21 69
M 21 12 12 16 64
F 29 13 10 23 69
M 14 6 6 14 53
F 14 11 5 21 69
M 17 13 12 22 64


40
21
34.8
24.1
18.2
19
54.1
29.4
44.7
56
18
27
44.1
59.8
21.4
22.1
34
21
27.2
21.8
24
26.8
43.1
59.3
34.9
53
43
24









CHAPTER 3
RESULTS

Recall that in this study, I was interested in examining the differences between adults who

stutter and adults who do not stutter in speech fluency, speech reaction time, and responses to

priming across sentences varying in their syntactic structure. Several analyses were conducted to

examine differences between the two groups in (a) the overall frequency of disfluencies they

produced, (b) the frequency of disfluencies produced pertaining to sentences of varying syntactic

structure, (c) the effect of priming on the frequency of disfluencies they produced, (d) the extent

to which syntactically primed structures were incorporated into responses, (e) overall mean

SRTs, and (f) mean SRT for each of several sentence types.

This chapter on results is organized into three main subsections that are based on the

analyses performed to address the above points. The first subsection deals with the fluency

analyses. In this section, results pertaining to the statistical analyses that address points (a) and

(b) above are described. The second subsection deals with syntactic priming analyses. In this

section, results pertaining to the statistical analyses used to address points (c) and (d) above are

described. The third subsection deals with the speech reaction time analyses. In this section,

results pertaining to the statistical analyses used to address points (e) and (f) above are described.

Fluency-Related Results

Three analyses were used to answer the fluency-related questions. The analyses were based

on the three sentence type experiments in the present study (i.e., Transitive, Dative, and Two-

clause). The first analysis examined the frequency and type of speech disfluencies produced by

each group during the Transitive experiment across the different prime types (i.e., active,

locative, and passive primes). The second analysis examined the frequency and type of speech

disfluencies produced by each group during the Dative experiment across prime types (i.e.,









double-obj ect, obj ect-complement, and prepositional obj ect primes). The third analysis examined

the frequency and type of speech disfluencies produced by each group during the Two-Clause

experiment across the different prime types (i.e., conj oined, center-embedded, and right-

embedded primes). A fourth analysis was conducted to compare the frequency and type of

speech disfluencies produced by each group across the three experiments (i.e., Transitive,

Dative, and Two-Clause). In the remainder of this section, results of these four analyses are

presented.

Transitive Experiment

Figure 3-la shows the frequency with which 'repetitions and prolongations' and 'other'

types of disfluencies were produced by participants in the two groups in the Transitive

experiment. As can be seen, overall, the PWS seemed to produce more 'repetitions and

prolongations' across the three examined transitive conditions (i.e., active, locative, passive) than

the PWNS.

To examine fluency in the Transitive experiment, a 2 (Group) x 2 (Disfluency Type) x 3

(Prime Type) multivariate analysis of variance (MANOVA) was conducted, with Group (PWNS,

PWS) as the between-subj ects factor and Disfluency Type ('repetitions and prolongations',

'other') and Prime Type (Active, Locative, Passive) as within-subj ects factors. The dependent

variable was the frequency of 'repetitions and prolongations' and 'other' types of disfluencies

the participants produced per Transitive prime type.

The main effect for the within-subj ects factor Disfluency Type was significant (F (1, 26) =

16.562, p < 0.001, r12 = 0.3 89) and the Group x Disfluency Type interaction was also significant

(F (1, 26) = 24. 143, p < 0.001, r12 = 0.481) indicating that the difference between the frequency

of 'repetitions and prolongations' and the frequency of 'other' disfluencies per response for one

study group were different from those frequencies for the other study group. The main effect of










the between-subj ects factor, Group was also significant (F (1,26) = 34.50, p < 0.001, rl = 0.57)

indicating that the PWS produced significantly more overall speech disfluencies in the transitive

responses (M~= 0.80) than the PWNS (M~= 0. 11).

To further examine the Group x Disfluency interaction, post-hoc paired-samples t tests

were run comparing the within-subj ects factor (i.e., frequency of 'repetitions and prolongations'

and 'other' types of disfluencies) in the following experimental condition pairs to determine

which were significantly different:

* 'Repetitions and prolongations' produced by PWNS versus 'repetitions and prolongations'
produced by PWS

* 'Other' disfluencies produced by PWNS versus 'other' disfluencies produced by PWS

A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/2 t tests =

.025). The results of one comparison was significant indicating that the PWNS produced

significantly fewer 'repetitions and prolongations' per response following the transitive primes

examined in the present study (M~= 0.067) than PWS (M~= 1.28) (t (26) = -5.864, two-tailed p <

0.001). The results of the other comparison was not significance (t (26) = -1.898, two-tailed p =

0.069) indicating that although PWNS produced fewer 'other' types of disfluencies in the

transitive responses following the transitive primes examined in the present study than PWS, the

difference in the frequency of 'other' types of disfluencies was not significant between the two

groups.

None of the results of the Prime Type analyses for the transitive experiment were

significant, indicating that, for all participants, the transitive prime types examined in the present

study did not seem to affect the frequency of 'repetitions and prolongations' versus 'other '

disfluency types produced in responses to those primes.









In summary, the results of the Transitive experiment analyses indicated that PWS

produced significantly more 'repetitions and prolongations' than PWNS. The results also

indicated that participants in the two study groups produced a comparable number of 'other'

disfluencies in responses following the transitive primes examined in the present study. In

addition, the results indicated that for all participants, the frequency and type of speech

disfluencies produced in the transitive responses did not seem to be affected by whether or not

the priming sentences preceding those responses exhibited an active, locative, or passive

transitive sentence structure.

Dative Experiment

Figure 3-1b shows the frequency with which 'repetitions and prolongations' and 'other'

types of disfluencies were produced by participants in the two groups in the Dative experiment.

As can be seen, overall, the PWS seemed to produce more 'repetitions and prolongations' across

the three examined dative conditions (i.e., double-object, object-complement, prepositional-

obj ect) than the PWNS.

To examine fluency in the Dative experiment, a 2 (Group) x 2 (Disfluency Type) x 3

(Prime Type) MANOVA was conducted, with Group (PWNS, PWS) as the between- subj ects

factor and Disfluency Type ('repetitions and prolongations', 'other') and Prime Type (Double-

object, Object-complement, Prepositional object) as the within-subjects factors. The dependent

variable was the frequency of 'repetitions and prolongations' and 'other' types of disfluencies

the participants produced per Dative prime type.

Consistent with results from the Transitive experiment, the main effect for the within-

subj ects factor, Disfluency Type, was significant (F (1, 26) = 19.912, p < 0.001, r12 = 0.434) and

the Group x Disfluency Type interaction was also significant (F (1, 26) = 21.63 8, p < 0.001, r12

0.454), indicating that the difference between the frequency of 'repetitions and prolongations'









and the frequency of 'other' disfluencies per response type for one study group were different

from those frequencies for the other study group. The main effect of the between-subjects factor,

Group was also significant (F (1, 26) = 32.863, p < 0.001, rl = 0.55) indicating that the PWS

produced significantly more overall speech disfluencies for their dative responses (M~= 0.90)

than the PWNS (M~= 0. 11).

To further examine the Group x Disfluency interaction, post-hoc paired-samples t tests

were run comparing the within-subj ects factor (i.e., frequency of 'repetitions and prolongations'

and 'other' types of disfluencies) in the following experimental condition pairs to determine

which were significantly different:

* 'Repetitions and prolongations' produced by PWNS versus 'repetitions and prolongations'
produced by PWS

* 'Other' disfluencies produced by PWNS versus 'other' disfluencies produced by PWS

A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/2 t tests =

.025). The results of one comparison was significant indicating that the PWNS produced

significantly fewer 'repetitions and prolongations' (M~= 0. 100) per dative primes than PWS (M~

1.54) (t (26) = -5.577, two-tailed p < 0.001). The results of the other comparison was not

significant indicated that following the dative primes examined in the present study, participants

in the two study groups produced a comparable number of 'other' types of disfluencies per

dative response (PWNS M~= 0. 12, PWS M~= 0.26) (t(26) = -2.017, two-tailed p = 0.065).

None of the results of the Prime Type analyses were significant indicating that for all

participants, the dative prime types examined in the present study did not seem to affect the

frequency of 'repetitions and prolongations' versus 'other' disfluency types produced in

responses to those primes.









In summary, the results of the Dative experiment analyses indicated that PWS produced

significantly more 'repetitions and prolongations' than PWNS. The results also indicated that

participants in the two study groups produced a comparable number of 'other' types of

disfluencies. Additionally, the results indicated that for all participants, the frequency and type of

speech disfluencies produced in the dative responses did not seem to be affected by whether or

not the priming sentences preceding those responses exhibited a double-object, obj ect-

complement, or prepositional object dative sentence structure.

Two-Clause Experiment

Figure 3-1e shows the frequency with which 'repetitions and prolongations' and 'other'

types of disfluencies were produced by participants in the two groups in the Two-Clause

experiment. As can be seen, overall, and similar to the results of the two previous analyses,

overall, the PWS seemed to produce more 'repetitions and prolongations' across the three

examined two-clause conditions (i.e., conj oined, center-embedded, right-embedded) than the

PWNS.

To examine fluency in the Two-Clause experiment, a 2 (Group) x 2 (Disfluency Type) x 2

(Prime Type) MANOVA was conducted, with Group (PWNS, PWS) as a between-subj ects

factor and Disfluency Type ('repetitions and prolongations', 'other') and Prime Type (Right

embedded, Center embedded, and Conj oined) as the within-subj ects factors. The dependent

variable was the frequency of 'repetitions and prolongations' and 'other' types of disfluencies

per response type.

Similar to the results of the Transitive and Dative experiments analyses, the main effect for

the within-subj ects factor Disfluency Type was significant (F (1, 26) = 34.809, p < 0.001 r12

0.572) and the Group x Disfluency Type interaction was also significant (F (1, 26) = 31.603, p <

0.001, r12 = 0.549) indicating, once again, that the difference between the frequency of










'repetitions and prolongations' and the frequency of 'other' disfluencies per response type for

one study group were different from those frequencies for the other study group. The main effect

of the between-subj ects factor, Group was also significant (F (1, 26) = 32.86, p < 0.001, rl =

0.55) indicating that the PWS produced significantly more overall speech disfluencies their two-

clause responses (M~= 0.87) than the PWNS (M~= 0.20).

To further examine the Group x Disfluency Type interaction, post-hoc paired-samples t

tests were run comparing the within-subj ects factor (i.e., frequency of 'repetitions and

prolongations' and 'other' types of disfluencies) in the following experimental condition pairs to

determine which were significantly different:

* 'Repetitions and prolongations' produced by PWNS versus 'repetitions and prolongations'
produced by PWS

* 'Other' disfluencies produced by PWNS versus 'other' disfluencies produced by PWS

A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/2 t tests =

.025). Similar to the results of the two previous analyses, the results of one comparison was

significant indicated that the PWNS produced significantly fewer 'repetitions and prolongations'

per two-clause responses examined in the present study (M~= 0.21) than PWS (M~= 1.46) (t (26)

= -6.249, two-tailed p < 0.001) and that the number of 'other' types of disfluencies per response

was similar between the two groups.

As with the two other experiments, none of the results for the Prime Type analyses were

significant indicating that for all participants, the two-clause prime types examined in the present

study did not seem to affect the frequency of 'repetitions and prolongations' versus 'other'

disfluency types produced in responses to those primes.

In summary the results of the Two-Clause experiment analyses indicated that PWS

produced more 'repetitions and prolongations' than PWNS. The results also indicated that










participants in the two study groups produced a comparable number of 'other' types of

disfluency in responses following the two-clause primes examined in the present study. In

addition, the results indicated that for all participants, the frequency and type of speech

disfluencies produced in the two-clause responses did not seem to be affected by whether or not

the priming sentences preceding those responses exhibited a right embedded, center embedded,

or conj oined two-clause sentence structure.

Fluency Across Picture Types

Before describing this analysis in detail, it is worth mentioning that although the present

analysis and three above analyses may seem similar, the present analysis examined the overall

effect of picture type (i.e., whether a picture was designed to elicit a Transitive, Dative, or Two-

Clause sentence) on the frequency and type of disfluencies between the two groups. In contrast,

the three previous analyses examined the effect of a given prime type (e.g., Transitive: Active,

Locative, Passive) on the frequency and type of speech disfluencies within that experiment.

Figure 3-2 shows the frequency with which 'repetitions and prolongations' and 'other'

types of disfluencies were produced by participants in the two groups in responses to different

picture types across the three experiments (i.e., Transitive, Dative, and Two-Clause). As can be

seen, overall, the PWS produced somewhat more 'repetitions and prolongations' (M~= 1.47) and

more than the PWNS (M~= 0. 12) across the three examined picture types.

To examine participants' fluency across the picture types used in the three experiments, a 2

(Group) x 2 (Disfluency Type) x 3 (Picture Type) MANOVA was conducted, with Group

(PWNS, PWS) as a between-subj ects factor and Disfluency Type ('repetitions and

prolongations', 'other') and Picture Type (Transitive, Dative, Two-clause) as within-subjects

factors. The dependent variable was the frequency of each type of disfluency per trial.









The main effect of the within-subj ects factor, Disfluency Type, was significant (F (1, 26) =

23.009, p < 0.001, r12 = 0.469) indicating that for all participants, the frequency of 'repetitions

and prolongations' was different from the frequency of 'other' types of disfluencies. The Group

x Disfluency Type interaction was also significant (F (1, 26) = 24.893, p < 0.001, r12 = 0.489)

indicating that that the frequency of 'repetitions and prolongations' versus 'other' types of

disfluencies for one study group was different from that frequency for the other study group. The

main effect of the between-subj ects factor Group was also significant (F (1, 26) = 3 8.613, p <

0.001, r12 = 0.598) indicating that overall, the PWS produced significantly more speech

disfluencies per response (M~= 0.89) than the PWNS (M2= 0. 13).

To further examine the Group x Disfluency Type interaction, post-hoc paired-samples t

tests were run comparing the within-subj ects factor (i.e., frequency of 'repetitions and

prolongations' and 'other' types of disfluencies) in the following experimental condition pairs to

determine which were significantly different:

* 'Repetitions and prolongations' versus 'other' disfluencies produced by PWNS

* 'Repetitions and prolongations' versus 'other' disfluencies produced by PWS

* 'Repetitions and prolongations' produced by PWNS versus 'repetitions and prolongations'
produced by PWS

* 'Other' disfluencies produced by PWNS versus 'other' disfluencies produced by PWS

A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/4 t tests =

.0125). The results indicated that the PWS produced significantly more 'repetitions and

prolongations' (M~= 0. 13) than 'other' disfluencies (M~= 0.29) per response following the picture

types used in the present study, (t (13) = 5.860, two-tailed p < 0.001). The comparison for the

PWNS was not significant indicating that the PWNS produced a comparable number of

'repetitions and prolongations' and 'other' types of disfluencies in responses following the










picture types used in the present study. In addition, the results indicated that when comparing the

two groups in terms of the types of disfluencies they produced, the PWNS produced significantly

fewer 'repetitions and prolongations' (M~= 0. 12) per response than the PWS (M~= 1.47) (t (13) =

-7. 139, two-tailed p < 0.001), but the difference between the groups in the numbers of 'other'

disfluencies was not statistically significant (PWNS M~= 0. 14, PWS M~= 0.31).

The main effect for the within-subjects factor, Picture Type, approached significance (F (2,

52) = 3.007, p = 0.058) and the Picture Type x Group interaction (F (2, 52) = 1.295, p = 0.283)

was not significant, indicating that although the PWS produced more speech disfluencies than

PWNS, all participants seemed to produce overall fewer speech disfluencies per response

following two-clause primes (M~= 0.45) than speech disfluencies in responses following both

transitive (M~= 0.54) and dative (M~= 0.54) primes.

There was a significant Disfluency Type x Picture Type interaction (F (2, 52) = 3.377,=

0.042, 12 = 0. 115). However, the Disfluency Type x Picture Type x Group interaction was not

significant (F (2, 52) = 1.446, p = 0.245) indicating that although picture type seemed to affect

fluency, it did not seem to affect the type of speech disfluency differently between study groups.

To further examine the Disfluency Type x Picture Type interaction, post-hoc paired-

samples t tests were run comparing the within-subj ects factor (i.e., frequency of 'repetitions and

prolongations' versus 'other' types of disfluencies) across the three prime types to determine

which were significantly different:

* 'Repetitions and prolongations' in responses following Transitive versus Dative pictures

* 'Repetitions and prolongations' in responses following Transitive versus Two-Clause
pictures

* 'Repetitions and prolongations' in responses following Dative versus Two-Clause pictures

* 'Other' disfluencies in responses following Transitive versus Dative pictures










* 'Other' disfluencies in responses following Transitive versus Two-Clause pictures

* 'Other' disfluencies n responses following Dative versus Two-Clause pictures

A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/6 t tests =

.008). The result of only one paired comparison was significant, indicting that participants in the

two study groups produced significantly fewer speech disfluencies in responses to the Transitive

picture types (M~= 0.45) than in responses to Two-Clause picture types (M~= 0.54) (t (27) = -

2.945, two-tailed p = 0.007).

In summary the results of the Picture Type analyses indicated that when responding to the

transitive, dative, and two-clause pictures examined in the present study, PWS produced

significantly more speech disfluencies than PWNS. Additionally, participants in the two study

groups produced significantly more overall disfluencies in responses following the two-clause

primes than in responses following the transitive and dative primes.

Priming Analyses

Similar to the procedures used in the fluency analysis described above, three analyses were

used to answer the priming related questions. Recall that these analyses were based on the three

experimental conditions (i.e., Transitive, Dative, and Two-clause). The first analysis examined

the percentage with which passive responses were produced following active, locative, and

passive primes in the Transitive experiment. The second analysis examined the percentage with

which prepositional obj ect dative responses were produced following double-object, obj ect-

complement, and prepositional obj ect dative primes in the Dative experiment. The third analysis

examined the percentage with which embedded responses were produced following right-

embedded, center-embedded and conj oined responses in the Two-Clause experiment. The

remainder of this section presents the results of these three analyses.









Transitive

Figure 3-3a shows the percentage with which passive responses were produced by

participants in the two groups following the presentation of active, locative, and passive primes.

As can be seen, the observed percentage of passive responses overall and passive responses to

passive primes for the PWNS was somewhat higher (M~= 23%) than it was for the PWS (M~=

12%). As will be shown in this section, the results of the statistical analyses examining these

overall and specific differences in the percentage of passive responses between the groups were

not statistically significant.

Priming-related analyses for the Transitive experiment were conducted using a 2 (Group) x

3 (Prime Type) mixed model analysis of variance (ANOVA). Group (PWNS, PWS) was the

between-subj ects factor, and Prime Type (Active, Locative, and Passive) was the within-subj ects

factor. The dependent variable was the percentage of sentences produced as full passives (i.e.,

the responses had to include a form of to be plus a by phrase) in the Transitive experiment.

Results showed no main effect for the within-subjects factor Prime Type (F (2, 52) = 0.845, p =

0.43 5), no Prime Type x Group interaction (F (2, 52) = 0.498, p = 0.611i) and no main effect for

the between-subj ects factor Group (F (1, 26) = 0.813, p = 0.375) indicating that participants in

the two study groups produced a comparable number of passive responses after active, locative,

and passive primes.

Dative

Figure 3-3b shows the percentage with which prepositional dative responses were

produced by participants in the two groups following the presentation of a double-obj ect, obj ect-

complement, and prepositional-obj ect dative primes. As can be seen, the observed percentage of

prepositional-obj ect responses to other dative responses for the PWS was somewhat higher for

two of the dative prime conditions ( Prepositional-obj ect responses following double-obj ect









primes M~= 43% and prepositional-obj ect responses following prepositional-object primes M =

54%) than they were for the PWNS (M~= 18% and 49% respectively). and prepositional-obj ect

responses following prepositional-obj ect primes M = 54%). As will be shown in this section, the

results of a number of the statistical analyses examining these overall and specific differences in

the percentage of prepositional-obj ect responses between the groups were statistically significant

Priming-related analyses for the Dative experiment were conducted using a 2 (Group) x 3

(Prime Type) mixed ANOVA. Group (PWNS, PWS) was the between-subjects factor, and Prime

Type (Double-Obj ect, Obj ect-Complement, Prepositional-Obj ect dative) was the within-subj ects

factor. The dependent variable was the percentage of prepositional obj ect dative responses in the

Dative experiment. Results showed a significant main effect for the within-subj ects factor, Prime

Type, (F (2, 52) = 9.455, p < 0.001, r12 = 0.267), and a Prime Type x Group interaction (F (2,

52) = 3.914, p = 0.026, r12 = 0. 131) indicating that participants in one study group produced more

prepositional-obj ect dative responses following at least one prime type than participants in the

other study group.

To further examine the main effect of Prime Type, post-hoc paired-samples t tests were run

comparing the within-subj ects factor (i.e., the percentage of prepositional-obj ect dative

responses) in the following priming type pairs to determine which were significantly different:

* Double-obj ect versus obj ect-complement primes
* Double-obj ect versus prepositional-obj ect responses
* Obj ect-complement versus prepositional-obj ect responses

A Bonferroni-corrected alpha was used to control Type I error (alpha levels = .05/3 t tests

=.016). The results indicated that all participants produced significantly more prepositional-

obj ect responses following prepositional-obj ect primes (M~= 52%) than following obj ect-










complement primes (M~= 23%) (t(27) = 4. 191, two-tailed p < 0.001) and following double-

obj ect primes (M~= 33%) (t(27) = 4. 191, two-tailed p = 0.016)

To further examine the Prime Type x Group interaction, post-hoc paired-samples t tests

were run comparing the within-subj ects factor (i.e., the percentage of prepositional obj ect dative

responses) in the following priming type pairs to determine which were significantly different:

* Responses produced by PWNS following double-obj ect versus obj ect-complement primes

* Responses produced by PWNS following double-obj ect versus prepositional-obj ect primes

* Responses produced by PWNS following obj ect-complement versus prepositional-obj ect
primes

* Responses produced by PWS following double-obj ect versus obj ect-complement primes

* Responses produced by PWS following double-obj ect versus prepositional-obj ect primes

* Responses produced by PWS following obj ect-complement versus prepositional-obj ect
primes

* Responses produced by PWNS versus PWS following double-obj ect primes

* Responses produced by PWNS versus PWS following obj ect-complement primes

* Responses produced by PWNS versus PWS following prepositional-obj ect primes

A Bonferroni-corrected alpha was used to control Type I error (alpha levels = .05/9 t tests

=.005). Pertaining to the PWNS, none of the results of the paired comparisons were significant.

In detail, although the PWNS produced relatively fewer prepositional-obj ect responses following

double-obj ect primes (18%) than following prepositional-obj ect primes (49%), the difference

between the two responses was not statistically significant (t (13) = -2.440, two-tailed p = 0.030).

Similarly, and although the PWNS produced fewer prepositional-obj ect responses following

obj ect-complement primes (3 0%) than following prepositional-obj ect primes, the difference

between the two responses for the PWNS was not statistically significant (t(13) = -2.181, two-

tailed p = 0.048). Thus the results indicated that although the PWNS produced more










prepositional-obj ect responses following prepositional-obj ect primes and following obj ect-

complement primes than following double-obj ect primes, the percentage of prepositional-object

dative responses produced by the PWNS was not significantly affected by the type of the dative

primes preceding those responses. Alternately, the results of only one paired-comparison for the

PWS were significant. Namely, PWS produced significantly fewer prepositional-obj ect

responses following obj ect-complement primes (16%) than prepositional-obj ect responses

following prepositional-obj ect primes (54%) (t (13) = -3.762, two-tailed p = 0.002). The result of

another paired-comparison (prepositional-obj ect responses following double-obj ect primes

versus following obj ect-complement primes) only approached significance (t (13) = 3.25 1, two-

tailed p = 0.006). Thus, for the PWS, the percentage of prepositional-obj ect responses was

affected by the type of dative prime preceding those responses. Specifically, the PWS produced

the significantly more prepositional-obj ect responses following prepositional-object primes

(54%) than following obj ect-complement primes (16%). Although the PWS seemed to produce a

more prepositional-obj ect responses following prepositional-obj ect than following double-obj ect

primes, the difference between the frequencies of prepositional-obj ect responses for this

examined comparison was not statistically significant. In terms of comparisons between the two

groups, none of the results of the paired-tests were significant indicating that although there were

differences between the two groups in the frequency of prepositional-obj ect responses following

the examined dative primes, the frequency of prepositional-obj ect responses following the

different dative primes examined in the present study was comparable between the two study

groups.

In summary, the results of the priming type in the Dative experiment indicated that overall,

participants tended to produce more prepositional-obj ect responses following prepositional-










obj ect primes than any other dative primes. The results also indicated differences between the

two groups in the frequency of prepositional-obj ect responses based on the structure of the

prime. Specifically for the PWNS, the type of dative prime did not seem to significantly affect

the probability using the prepositional-obj ect prime in the response. In addition, the PWNS

appeared to produce more prepositional-obj ect responses following prepositional-obj ect and

obj ect-complement primes than following double-obj ect primes, although none of the differences

were statistically significant. The PWS on the other hand tended to produce more prepositional-

obj ect responses following double-obj ect primes (the frequency was not significant compared to

the frequencies following the other primes), and significantly more such responses following

prepositional-obj ect primes than following obj ect-complement primes.

Two-Clause

Figure 3-3c shows the percentage with which various types of responses were produced by

participants in the two groups following presentation of a two-clause prime sentence. As can be

seen, the observed percentage of responses with embedded clauses was different between the two

groups; the PWNS produced complex sentences more frequently (M~= 61%) than the PWS (M~=

42%). As will be shown in this section, the results of the statistical analyses examining these

overall and specific differences in the percentage of embedded responses between the groups

were not statistically significant

Priming-related analyses for the Two-Clause experiment were conducted using a 2 (Group)

x 3 (Prime Type) mixed model ANOVA. Group (PWNS, PWS) was the between-subj ects factor,

and Prime Type (Conj oined, Right-Embedded, and Center-Embedded) was the within-subj ects

factor. The dependent variable was the percentage of complex sentence responses in the Two-

Clause experiment. Results showed no significant main effect for the within-subj ects factor,

Prime Type (F (2, 52) = 0.548, p = 0.581), and no Prime Type x Group interaction (F (2, 52) =









0.788, p = 0.460). The main effect for the between-subj ects factor Group approached

significance (F (1, 26) = 3.896, p= 0.059), indicating a variable response pattern across

participants for embedded clauses.

Summary of the Priming Analyses

In the Transitive experiment, the results indicated that overall, participants produced a

comparable number of passive responses following active, locative, and passive priming

sentences. In the Dative experiment, the results indicated that there were differences between the

two groups in the percentage of prepositional-obj ect responses based on the type of the dative

prime. Specifically and for the PWS, significantly more prepositional-obj ect responses were

produced following prepositional-obj ect primes than following obj ect-complement primes.

Finally, in the Two-Clause experiment, the results indicated that although their was a trend

toward PWNS producing more embedded responses than PWS, prime type (i.e., conj oined

sentences versus sentences with embedded clauses) did not affect the type of sentence produced

by either group.

SRT Analyses

Before describing the analyses used to examine SRT, it is worth mentioning that the

original plan was to look at SRT within the response types associated with each experiment (i.e.,

Transitive experiment: active, locative, and passive response types; Dative experiment: double-

object, object-complement, and prepositional object response types; Two-Clause experiment:

conj oined, right-embedded, and center-embedded response types), in addition to the congruency

of these responses with the priming sentence (e.g., frequency of passive responses following a

passive priming sentence). Recall also that only fluent responses were included in the SRT

analysis. For the purposes of this analysis, "fluent responses" were defined as those responses for

which at least the first noun phrase contained no instances of disfluency (i.e., responses that










followed the more "lenient" fluency definition described in the Methods chapter). When the

fluency-related criterion (i.e., including only responses for which NP 1 was produced fluently)

was applied across the categories of response types, many of the response type cells, particularly

for PWS, contained no data points. Consequently, data were collapsed and examined together

within each of the three experiments. Collapsing the data in this manner did not allow for

examination of SRT in terms of priming. Thus, the results reported in this subsection are for SRT

across the three prime types regardless of whether or not the responses matched the structure of

the exact the prime. Specifically, in the Transitive experiment, overall SRTs for responses that

exhibited a transitive sentence structure were examined without detailed examination and

comparison of SRTs for active versus locative versus passive responses. That is, overall SRTs

for the Transitive prime type were examined without comparing SRTs across the specific active,

locative, or passive transitive primes. Similarly, in the Dative experiment, overall SRTs for

responses that exhibited a dative sentence structure were examined regardless of the priming

condition in which they occurred. That is, overall SRTs for the Dative prime type were examined

without comparing SRTs across the specific double-object, object-complement, or prepositional-

object dative primes. Similarly, and in the Two-Clause experiment, overall SRTs for two-clause

for responses were examined regardless of the prime condition in which they occurred. That is,

SRTs for the Two-Clause prime type were examined without comparing SRTs across the

specific conj oined, right-embedded, or center-embedded two-clause primes.

Figure 3-4 shows the mean and standard deviations for SRT for response types across the

three experiments (i.e., responses that were transitive in the Transitive experiment, responses that

were dative in the Dative experiment and responses that were two-clause in the Two-Clause

experiment). As can be seen, when the responses exhibited a transitive sentence structure, SRTs









for the PWS were somewhat shorter (M~= 2274.69 ms) than SRTs for PWNS (M~= 2372.97 ms).

A similar difference and direction of the difference in SRT between the two groups can be also

seen when responses exhibited a dative sentence structure (PWS M~= 2404.25 ms, PWNS M~=

2604.89 ms) and a two-clause sentence structure (PWS M~= 263 8.41 ms, PWNS M~= 2780.76

ms). As will be shown, the results of the statistical analyses described in this section revealed

that these observed differences between the two groups were not statistically significant.

Two questions were asked to answer the study questions pertaining to SRT in relation to

speaker group and response type. The first question considered whether there was a difference

between PWS and PWNS in overall SRT. The second question asked if response type (i.e.,

transitive, dative, two-clause) affected speech reaction times of PWS and PWNS similarly.

To determine whether there were significant differences between the two groups in SRT, in

addition to any possible difference between the groups in mean SRT for the different response

types, a 2 (Group) x 3 (Response Type) repeated measures ANOVA was used. Group (PWNS,

PWS) was the between-subj ects factor and Response Type (responses that were transitive in the

Transitive experiment, responses that were dative in the Dative experiment, and responses that

were two-clause in the Two-Clause experiment) was the within-subj ects factor. As stated above,

the analyses included entirely fluent responses in addition to ones that had a fluent first noun

phrase. The puerility assumption was tested but was not met (X2 (2) = 11.615, p = 0.003). Thus,

the reported statistics are for the Huynh-Felt correction. There was a significant main effect for

the within-subj ects factor, Response Type, (F (1.58, 41.13) = 14.654, p < .001, r12 = .360).

However, there was no significant main effect for the between-subj ects factor, Group (F (1, 26)

= 0.302, p = 0.587), and no significant Group x Response Type interaction, (F (1.58, 41.13) =










.259, p = 0.721). Thus, for all participants, there was a difference in SRT in the responses they

produced for at least one of the response types examined in the present study.

To further examine the main effect of Response Type, post-hoc paired-samples t tests

were run comparing the within-subj ects factor (i.e., SRTs for transitive responses in the

Transitive experiment, dative responses in the Dative experiment, and two-clause responses in

the Two-Clause experiment) in the following response type pairs to determine which were

significantly different:

* Transitive versus dative
* Transitive versus two-clause
* Dative versus two-clause

A Bonferroni-corrected alpha was used to control Type I error (alpha levels = .05/3 t tests

=.016). The results of all paired comparisons were significant. In detail, the results indicated that

and for all participants, when producing sentences in which at least the first noun phrase was

fluent, SRTs for sentences exhibiting a transitive sentence structure were faster (M~= 2323 ms)

than SRTs for sentences exhibiting a dative structure (M~= 2504 ms) (t (27) = -4.029, two-tailed

p < 0.001) and were also faster than SRTs for sentences exhibiting a two-clause structure (M~=

2709 ms) (t (27) = -4.770, two-tailed p < 0.001). In addition, SRTs for sentences exhibiting a

dative structure were faster (M~= 2504 ms) than SRTs for sentences exhibiting a two-clause

structure (M~= 2709 ms) (t (27) = -2.587, two-tailed p = 0.015).

In summary, the results of the SRT analyses indicated that for all participants, SRTs for

transitive responses were faster than those for dative responses and SRTs for dative responses

were faster than those for two-clause responses.

To address the possibility that response length (in words) may have affected SRT across

the three response types (i.e., Transitive, Dative, and Two-Clause) (See Figure 3-5 for the mean









and standard deviation for utterance length (in words) in the responses produced across the three

experiments), a 2 (Group) x 2 (Response Type) analysis of covariance (ANCOVA) was used.

Group (PWNS, PWS) was the between- subj ects factor, Response Type (transitive responses in

the Transitive experiment, dative responses in the Dative experiment, and two-clause responses

in the Two-Clause experiments) was the within-subj ects factor, and Response Length (in words)

was the covariate. The dependent variable was mean SRT for transitive, dative, and two-clause

responses. As with the preceding analysis, this analysis included entirely fluent responses in

addition to ones that had a fluent first noun phrase. The puerility assumption was tested but was

not met (X2 (2) = 11.425, p = 0.003). Thus, the reported statistics are for the Huynh-Felt

correction. There was no significant main effect for the within-subj ects factor Response Type (F;

(1.64, 40.99) = 2.470, p = 0. 107), no Response Type x Group interaction (F (1.64, 40.99) =

0.267, p = 0.723), no Response Type x Response Length interaction (F (1.64, 40.99) = 1.646, p

= 0.208), and no main effect of the covariate Response Length (F (1,260) = 0.067, p = 0.798).

Thus, for all participants, there were no differences in SRTs among the examined response types

when differences in response length across experiments were considered.










Figure 3-1 Mean number and standard error for 'repetitions and prolongations' (Rep.& Pro.) and
'other' types of disfluencies (Other) in responses produced by persons who do not
stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when the priming
sentences were (a) Transitive (Active, Locative, Passive), (b) Dative (Double-Obj ect
= D.- obj., Obj ect-complement = Obj.-comp., Prepositional-obj ect = Prep.-obj.), and
(c) Two-Clause (Conj oined, Right-embedded = R-emb., Center-embedded = C.-
emb .).




















Rep.& Other Rep.& Other Rep.& Other
Pro. Pro. Pro.

Active Passive Locative

Transitive Prime Type



(b) Dative Prime Effect on Fluency





-I ..a .


(c) Two-Clause Prime Effect on Fluency









Rep.& Other Rep.& Other Rep.& Other


(a) Transitive Prime Effect on Fluency


o



04




0 0
'

0
c,






. e
p a
0 0
M
F


1.6
1 2
S0.8
0.4















1.6
1 2
S0.8
0.4


O PW NS
SPWS
















o PWNS
HPWS


Rep.& Other
Pro.

Obj.-comp.


Dative Prime Type


a,






a &


O PWNS
HPWS


Pro.


Pro.


Pro.


Conjoined


R- embed.


C- embed.


Two-Clause Prime Type


Rep.& Other
Pro.

D.-obj.


Rep.& Other
Pro.

Prep.-obj.










Effect of Picture Type on Fluency



po~ 1.2 OPWNS
FL 0 H PWS
O 0.4

a8 Rep.& Other Rep.& Other Rep.& Other
Pro. Pro. Pro.

Transitive Dative Two-Clause

Picture Type

Figure 3-2. Mean number and standard error for 'repetitions and prolongations' (Rep.& Pro.)
and 'other' types of disfluencies (Other) in responses produced by persons who do
not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when the priming
pictures were (a) Transitive (b) Dative, and (c) Two-Clause.










Figure 3-3. Mean percentage and standard error for responses produced by persons who do not
stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when the priming
sentences were (a) Transitive (Active, Locative, Passive), (b) Dative (Double-Obj ect,
Obj ect-complement, Prepositional-obj ect), and (c) Two-Clause (Conj oined, Right-
embedded, Center-embedded).










(a) Effect of Transitive Prime Type on Response


30

8 20

~ O

M 1


0 PWNS
SPWS


Active Passive Locative

Prime Type


(b) Effect of Dative Prime Type on Response


-2 m70
0 8 60
0 0 50
A 40
"oFL 30
og~ 20
0~ 0


0 PWNS
HPWS


Double- obj ect. Obj ect- Prepositional-
complement object
Prime Type

(c) Effect of Two-Clause Prime Type on Response


O PWNS
mPWS


Conjoined Right-embedded Center-embedded
Prime Type













*


Effect of Response Type on SRT


3500


3000 -


2500


2000'


O PWNS
HPWS


Transitive


Dative


Two-Clause


Response Type

Figure 3-4 Mean and standard error for speech reaction time (SRT) for responses produced by
persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14)
when those responses were (a) Transitive in the Transitive experiment, (b) Dative in
the Dative experiment, and (c) Two-Clause in the Two-Clause experiment.




Response Length in Words


O PWNS
HPWS


Transitive Dative Two-Clause


Response Type

Figure 3-5 Mean and standard deviation for number of words produced by persons who do not
stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) for (a) Transitive
responses in the Transitive experiment, (b) Dative responses in the Dative
experiment, and (c) Two-Clause responses in the Two-Clause experiment.









CHAPTER 4
DISCUSSION

Three main questions were asked in this study: (1) Does syntactic structure affect speech

fluency in persons who stutter and persons who do not stutter? (2) Does syntactic structure affect

speech reaction time in persons who stutter and persons who do not stutter? (3) Does syntactic

priming affect speech reaction time in persons who stutter and persons who do not stutter? and

(4) Does syntactic priming affect speech fluency in persons who stutter and persons who do not

stutter?

These questions were motivated by theories which suggest a relationship between

stuttering and difficulties in the retrieval and/or formulation of syntactic information during

speech production. This study was designed to test these theories and to extend prior research by

examining the relationship between syntactic structure, speech fluency, and speech reaction time

in persons who stutter. Fourteen adults who stutter and fourteen adults who do not stutter

participated in the study. Responses were elicited using a syntactic priming paradigm during

which participants repeated sentences and then described pictures depicting actions that can be

described using transitive, dative, and two-clause sentence forms. In this chapter, I remind the

reader of the main study findings and discuss the general implications for those Eindings in light

of previous research. Accordingly, the reminder of this chapter is designed as follows. In the first

section, the results of the fluency analyses and the fluency in relation to priming are presented

and discussed. In the second section, the results of the priming analyses are presented and

discussed. In the third section, the results of the SRT analyses and the SRT in relation to

response type are presented and discussed. In the fourth and Einal section, the limitations of the

present study in addition to suggestions for future research are presented.









The Effect of Syntactic Structure on Fluency

Recall that in the present study, the purpose of using the above presented syntactic priming

paradigm was to elicit responses reflecting a gradual increase in syntactic complexity. Namely,

as defined in the present study, transitive sentences (i.e., active, locative, passive) reflected a low

level of syntactic complexity, dative sentences (i.e., double-object, object-complement,

prepositional-obj ect) reflected an intermediate level of syntactic complexity, and two-clause

sentences (i.e., conj oined, right-embedded, center-embedded) reflected a high level of syntactic

complexity. The main finding from the fluency analyses indicated that although the PWS

produced significantly more speech disfluencies than the PWNS, the syntactic structure (i.e.,

syntactic complexity) of the priming sentences used in the present study did not seem to affect

fluency differently between the two study groups. This result is consistent with findings from

previous studies that have examined the effect of syntactic complexity upon the speech fluency

of adolescents and adults who stutter (e.g., Silverman and Ratner, 1997; Logan, 2001). This

result, however, is not consistent with a number of previous findings from studies involving

young children and adults who stutter. In those studies, results have pointed toward a significant

relationship between the syntactic complexity of a sentence and the fluency with which it is

spoken (e.g., Bernstein Ratner, 1997; Bosshardt, Ballmer, & De Nil, 2002; Logan and Conture,

1995).

It is interesting to consider the underlying factors for the difference in the effect of

syntactic complexity on speech production between earlier reports indicating an effect in

children who stutter and the results of the present study. One explanation pertains to the fact that

almost all the research done with children who stutter to explore the effect of varying syntactic

structure on speech production has examined this relationship in preschool and young school age

children. As suggested in a number of previous studies (e.g., Silverman and Ratner, 1997; Logan,










2001, Yaruss, 1999), the syntactic structures used in experiments with younger speakers might

be developmentally challenging to those children. For instance, complex syntax forms may take

young children a relatively long time to retrieve or assemble and this in turn could place added

stress on the speech motor system. An alternate possibility is that the children who stutter are

prone to making excessive syntactic formulation during the sentence formulation process. Any

such errors would seemingly have to occur prior to overt speech production, however, because

children who stutter have not been found to produce substantially more syntactic errors in their

spoken language than children who do not stutter. In any case, it is logical to assume that with

age and increased competency in one's native language, what might have been considered a

challenging linguistic form for a younger age group might not pose much challenge in

adolescence or adulthood. Along the same lines and within the framework of earlier presented

models of stuttering (e.g., the Demands and Capacities Model), increased syntactic complexity

could be more demanding on children's speech and language systems (as compared to adults)

and thus, increased percentage of speech disfluency in responses produced by children who

stutter could be viewed as resulting from inability of their language (and probably speech)

systems to handle such demands during ongoing speech. As Yaruss (1999) suggested, longer

and/or more syntactically complex sentences contain more information units than shorter and

less syntactically complex sentences. When processing the more syntactically complex sentence

structures, more such information units may need to be held in short-term memory until

decisions about the target utterance are made. As Yaruss explained, this in turn might exert more

demands on the processing system resulting in a higher percentage of disfluencies in the more

syntactically complex responses than in the less syntactically complex ones. Thus, it is









reasonable to suggest that from a developmental point of view, it is not surprising that syntactic

structures that have been reported as challenging to children are not as challenging to adults.

Although syntactactic complexity does not appear to affect overt fluency in AWS, it does

seem to affect other types of functions. For example, Bosshardt et al. (2002) reported that when

speakers in their study (i.e., adults who stutter and adults who do not stutter) were asked to

generate and produce sentences during a single-task (using two given nouns to generate a

sentence) versus dual-task paradigm (using two given nouns to generate a sentence and at the

same time decide if the two nouns are in the same noun category or if they rhyme) both adults

who do and do not stutter exhibited a reduced percentage of accurate rhyming and category

decisions. In addition, persons who stutter produced significantly fewer propositions (i.e., a

predicate and its argumentss) under the dual-task than persons who do not stutter. The results

were taken to indicate a "vulnerability" in the speech- production systems of persons who stutter

which is observed during "attention-demanding" tasks such as the secondary rhyme and category

decision tasks used in that study. Similarly, Cuadrado et al. (2003) reported both neural-

processing and response accuracy differences between persons who stutter and persons who do

not stutter during online grammaticality judgment tasks. Specifically, and when asked to decide

online (i.e., decide during or immediately following a response) whether the verbs in a

syntactically simple versus a syntactically complex sentence indicted a grammaticality

agreement, the percentage of accurate grammaticality judgments for persons who stutter was

significantly lower than for participants who do not stutter for grammaticality violations that

occurred in the more syntactically complex sentences. The results of the grammaticality

judgment task in addition to results of ERP measures collected during those tasks were taken by









Cuadrado et al. (2003) to indicate that "neural mechanisms reflecting postlexical analysis,

specifically for syntactic reanalysis, operate differently in (PWS)) (p.974).

It is worth mentioning that the study tasks in Bosshardt et al. (2002) involved generation

and production of the responses while in Cuadrado et al. (2003) no speech production was

involved required. Thus under tasks that are considered more demanding and even in the absence

of speech production, persons who stutter seem to exhibit difficulties in sentence planning and

production than persons who do not stutter. Thus, in summary, it seems that results of studies

examining overt speech fluency (e.g., frequency of disfluencies per response in relation to certain

syntactic forms) have shown a comparable performance between the PWS and the PWNS on the

syntactic tasks used in those studies. On the other hand, studies of more subtle aspects of

linguistic performance (e.g., those studies examining syntactic processing during sentence

generation through ERP examination) have shown an difference between persons who stutter and

persons who do not stutter.

The syntactic priming paradigm used in the present study could be viewed as more

challenging than mere repetition of given utterances because of the fact that it necessitates that a

participant generates a novel sentence instead of repeating the sentence he or she just heard (as

opposite to the data elicitation paradigm used in Logan (2003) study viewed above). In addition,

the paradigm used in the present study employed a recognition memory task which required the

participants to indicate whether or not they heard or saw the target sentences before within the

framework of the experiment and which served as a secondary or concurrent task. Based on these

two characteristics of the priming paradigm used, and although no significant effect of syntactic

structure was observed in the present study, I believe that the syntactic priming task used in the

present study elicited sentences was in some ways closer to what a speaker might produce in










everyday speaking contexts in that they required that a speaker to "creatively" produce a

sentence rather than merely produce a sentence they had just heard or read, as was the case in

studies by Silverman and Ratner (1997) and Logan (2001). Accordingly, the results of the

present study could be viewed as an addition to the increasing body of literature that argues

against a role for syntactic complexity independent of whatever effect that syntax may have

upon utterance length in the frequency with which adults who stutter produce overt disfluency.

Two other findings for the fluency analyses reported in the present study pertain to the

frequency and type of speech disfluencies produced (i.e., 'repetitions and prolongations' versus

'other 'types of disfluencies). The first finding indicated that overall, the PWS produced

significantly more 'repetitions and prolongations' per response than the PWNS. This finding is

not surprising based on the fact that (a) 'repetitions and prolongations' are considered central or

core behaviors in stuttering (see Bloodstein, 1995 for a detailed description), and (b) group

selection criterion were based on, among other considerations, the frequency with which

participants produced these types of disfluency in connected speech. The second finding

indicated that participants in the two study groups produced a comparable number of 'other'

types of disfluencies per response across the examined sentence types. This finding is consistent

with results from other studies that have compared the frequency of this disfluency class between

speaker who do and do not stutter. The overlap in frequency with which speakers from both

groups produced 'other' disfluencies such as interj sections and revisions in could reflect general

similarities between the two groups in (a) speech processing mechanisms, or (b) general

language formulation aptitude (e.g., lexical selection, syntactic formulation). Earlier studies have

shown that such disfluencies are common in typical speakers (see Bailey and Ferreira, 2003 for a

review) and, leaving aside instances when speakers who stutter use interj sections and revision as a









strategy for concealing 'repetitions and prolongations' from listeners, seem to be peripheral to

the fundamental impairment that characterizes development stuttering.

Another nonsignificant finding related to the fluency analyses warrants some discussion. In

the present study, PWNS produced a comparable number 'repetitions and prolongations' and

'other disfluencies in most of the experimental conditions. This finding was somewhat

unexpected because previous research on conversational fluency in PWNS has found that they

produce roughly twice as many 'other 'disfluencies as they do 'repetitions and prolongations'

(Bloodstein, 1995). Thus, this disfluency pattern for the PWNS suggests that sentence production

tasks used in the present were somewhat challenging for the PWNS.

The Effect of Picture Type on Fluency

As you may recall, all participants in the present study produced significantly more

disfluencies in responses following the two-clause primes than in responses following the

transitive primes. This finding is consistent with the results of Silverman and Ratner' s (1997)

study which reported an increase in the number of normal disfluencies (e.g., interj sections and

revisions) by adolescents who stutter and adolescents who do not stutter (ages 10-18 years)

(although in the present study, the reported significant increase was overall, i.e., for all types of

disfluencies regardless of whether they were 'repetitions and prolongations' or 'other' types of

disfluencies). In the Silverman et al. (1997) study, the significant increase in 'other' types of

disfluencies was observed in center-embedded responses (or structures defined in that study as

being "hardest" in terms of syntactic complexity) when compared to questions (defined as being

"easiest" in terms of syntactic complexity) and when compared to right-embedded responses

(defined as being "moderate" in terms of syntactic complexity). The authors also reported the

frequency of repetitions and prolongations (or as defined in that study "stuttered disfluencies")

was not significantly affected by syntactic complexity. In the present study, and as described on










the previous chapter, the frequency of both 'repetitions and prolongations' and 'other' types of

disfluencies seemed to significantly increase under conditions of increases syntactic complexity.

As explained in the Methods chapter, the transitive sentences and pictures used in the

present study each include two arguments (an agent and a patient). The dative sentences and

pictures on the other hand include three arguments (an agent, a direct patient, and a second

indirect patient). The two-clause sentences and pictures include four arguments (an agent and a

patient in the main clause, and an agent and a patient in the embedded clause or in the second

independent clause in the case of the conj oined primes). It can be argued that the more

arguments a response includes, the more complex that response is in terms of the computational

resources involved in generating such a response under time constraints similar to the ones used

in the present study. In terms of differences in frequency of disfluencies between the two-clause

and transitive responses examined in the present study, it is reasonable to suggest that the two-

clause responses impose greater memory loads than that imposed by the generation of transitive

and dative sentence forms. These memory loads might result from the fact that when generating

right-embedded responses as in "The woman pulled the man who pulled the dog," one 'agent'

has to be held in memory and when generating center-embedded responses as in "The woman

who pulled the man pulled the dog" two agents have to be held in memory to be matched with

their verbs.

The Effect of Syntactic Structure on Speech Reaction Time

Recall than in the only study that examined SRT in adults who stutter in relation to

syntactic complexity (i.e., Logan, 2003), the author reported that SRT was significantly longer

for the PWS than the nonstuttering controls for three of the four sentence types used in that

study. The first finding in the present study was that the overall SRT values for PWS were on

average 70.42 ms shorter than the overall SRT for PWNS (mean SRT for PWNS = 2667.20 ms









and for PWS = 2596.78 ms). This result is inconsistent with earlier Eindings which reported that

the overall mean SRTs for the PWS were longer (by approximately 132 ms) than SRTs for

PWNS (Logan, 2003) for the linguistic stimuli used in that study. This finding is also

inconsistent with previous findings indicating that the PWS tend to take more time than PWNS

to initiate vocal responses to non linguistic stimuli (see review in Logan, 2003). It is not possible

to definitely determine why the present findings are inconsistent with those previous findings. It

is possible that although the participant selection procedures used in the present study were

designed to ensure random selection, the group that actually participated might represent a

subgroup of persons who stutter whose speech production processes are not as affected by

linguistic and motoric planning demands as reported for participants in the earlier referenced

studies. A number of earlier investigations have implied that speakers who stutter are a

heterogeneous group and that generalizations about stuttering behaviors might not always valid

for all persons who stutter (Bloodstein, 1995). For example, Logan (2003) reported that not all

participants who stutter in his study exhibited longer SRTs compared to participants who did not

stutter. Thus, it reasonable to suggest that perhaps including many more participants in similar

future investigations might ensure a better representation of the population of speakers who

stutter and subsequently provide more details about the effect of both time constraints and

syntactic complexity on speech reaction time in those speakers.

The second finding of the SRT analyses in the present study indicated that for all

participants and when examining responses in which the first noun phrase was fluent, SRTs for

transitive response were significantly shorter than SRTs for dative responses which in turn were

significantly shorter than those for two-clause responses. However, when response length was









controlled (i.e., covaried) across experiments, the difference in SRT between the response types

was no longer observed.

Although the Einding pertaining to no differences between the two groups in speech

reaction time seems inconsistent with that reported in Logan (2003), the difference between the

reported results of the present study and Logan (2003) study could be attributed to a number of

factors. First, in his preparation of the study material, Logan elected to increase the syntactic

complexity of an utterance by generating sentences that differed in the elaboration of the subj ect

noun phrase prior the main verb in a given sentence. For example, two of the sentence forms

used by Logan (2003) were of (a) a Determiner + Adj ective + Adj ective + Noun as in "The long

and shiny car belongs to the girl" and (b) a Determiner + Noun + Relative Clause as in "The car

belongs to the girl who plays soccer." The structure of sentence form (a) employs pre-noun

modification (i.e., Determiner + Adj ective + Adj ective), while the structure of sentence form (b)

employs both pre and post noun modification (i.e., Determiner-pre, Relative Clause-post). As

mentioned above, Logan reported a significant difference between the two study groups in mean

SRT for three of the four sentence forms used in the study (among which are SRTs for responses

reflecting the structures of sentence forms (a) and (b) above). In the present study, the increase in

syntactic complexity was defined in terms of (a) the number of arguments described within a

given utterance, and (b) the number of independent and dependent clauses within that utterance.

In terms of the number of arguments within each sentence type, the transitive sentences included

two arguments, dative sentences included three arguments, and two-clause sentences included

four arguments. In terms of the number of independent and dependent clauses within each

sentence type, both transitive and dative sentences included only one independent clause and no

dependent clause, while the two-clause sentences included one independent and one dependent









clause. As explained in the Introduction and Methods chapters of the present study, earlier

studies have suggested that the more arguments a sentence includes the more challenging that

sentence is (see Shapiro, 1997 for an overview). The results of the SRT analyses in the present

study indicating an increase in SRT as the syntactic (and thematic or semantic) complexity of the

response increased seem to support these suggestions.

Although, as mentioned above, increased elaborations of the noun phrases (and in turn of

response length) within an utterance have been shown in earlier studies to affect speech reaction

times in both typical speakers and speakers who stutter (Ferreira, 1991; Logan, 2003) such

elaborations were not factored into sentence preparation in the present study. Recall that when

response length was controlled in the analyses, the difference in SRT between the response types

was no longer observed in the present study. Based on this finding, it could be argued that

although response length has been shown in previous studies to affect SRT, the effect of

syntactic structure cannot be ruled out as a contributing factor to differences in SRTs between

response types. It can be suggested that future research employing the syntactic priming

methodology could adopt alternative ways in defining the syntactic complexity such as

combining the definition of complexity used in the present study (i.e., the number of elements

within a given sentence in addition to number of independent and dependent clauses) and that

used by Ferreira (1991) and Logan (2003) (i.e., elaboration of noun-phrase elements within the

target utterance, thus increasing response length). Adopting this definition within the framework

of a syntactic priming paradigm might assist in providing additional views of any interactions

between increased syntactic complexity (and increased utterance length) on speech timing in

both speakers who do not stutter and speakers who stutter.









Two methodological points need to be considered when talking about the SRT analyses

reported in the present study. First, although I argue above that the syntactic priming paradigm

used in the present study involved generation of novel sentences by the participants, one could

not rule out the fact that the responses are not entirely novel. This is based on the fact that studies

employing this paradigm have consistently reported a tendency for participants to use many of

the structures in the priming sentences when they produce a response. Thus, for all responses

produced in the present study, a 'model' was always available to reuse. This in turn might have

reduced the time needed by all participants to generate the utterances and thus eliminated any

group differences that might have been observed had the structures in the target sentences been

truly novel. One way to overcome this could be by introducing a 'no prime' condition to the

priming study material. For example, instead of all the responses being preceded by a priming

sentence, some of target responses could be preceded by a 'no-prime' condition. In this case,

participants would have to generate, on their own and without the convenience of the modeled

structure in the priming sentence, a novel sentence structure and use it to describe that picture. In

this case, one could argue that differences between adults who stutter and adults who do not

stutter in initiating their sentences might be more pronounced. (Of course, this approach contains

one major drawback the inability to control for response type and response length.).

Nonetheless, the argument builds on the hypothesis that when asked to perform a reasonably

demanding task (such as one involving the generation and production of a response under time

pressure), subtle differences between the language and speech processing systems in persons

who stutter versus persons who do not stutter are observable when performing such a task.

Indeed, studies that have reported a difference in SRT between persons who stutter and persons

who do not stutter have shown no significant difference between the groups in SRT when a










prime was provided (similar to the report in the present study); however, such differences were

significant in the absence of a prime (e.g., Anderson and Conture, 2003).

Concerning the second methodological point (i.e., Data elicitation procedures and

instructions) recall that in the present study, and prior to the presentation of the study material

(i.e., priming pictures), participants were instructed to think of and say a sentence describing the

given pictures as soon as it appears on the computer screen. This instruction was given before

any material presentation and was only presented once to each participant. Recall also that on

average, the study task took about 45-60 minutes to Einish. The participants were not reminded of

that instruction anytime during the time course of material presentation. This was done to be true

to the methodological procedures used by Bock and colleagues in their priming studies and thus

be able to compare the results of the present study to the results of those priming studies. On the

other hand, in the previous studies that have reported an effect of syntactic structure on SRT (i.e.,

Logan, 2003), participants were specifically instructed and frequently reminded to initiate their

responses as fast as they could upon the presentation of an auditory cue (i.e., sound signal). It

could be argued that the instructions in Logan's study might be more likely to elicit between

groups differences in SRT than those used in the present study and that could be due to the fact

that time constraint was not given a central role in that participants were not instructed to

respond 'as fast as they can' and thus they might have taken their time to initiate their responses.

This in turn may have eliminated any group differences and thus resulted in not observing any

differences in SRT between groups in the present study. One suggestion for future research

examining SRT using the priming methodology with persons who stutter is to specifically

instruct the participants to initiate speech as fast as they can, and perhaps, even add reminders to

do that at different times during the time course of material presentation. This in turn might add









to the time pressure and provide addition information about how participants may behave under

both syntactic priming conditions and under continuous time pressure.

The Effect of Syntactic Priming on Response Type

The priming analyses in the present study yielded several findings. The first finding

suggested that priming methodology in general and syntactic priming in specific can be used to

elicit responses reflecting a variety of syntactic structures, including structures that are less

frequently observed such as datives and passives. This result was expected and is consistent with

earlier results with both typical speakers (e.g., Bock, 1986; Bock & Loebell, 1990; Smith &

Wheeldon, 2001) and with speakers who stutter (Anderson & Conture, 2004; Burger & Wijnen,

1999; Hartfield & Conture, 2006; Melnick, Conture, & Ohde, 2003; Pellowski & Conture, 2005;

Wijnen & Boers, 1994). As the reader may recall, the study material used in the present study

was adapted from material that was developed and has been used by Bock and colleagues for

almost twenty years. The sentence structures used in the present study were also ones that

previous syntactic priming studies have consistently reported to exhibit a priming effect. Thus, it

was not surprising that participants in the present study showed response patterns that were

similar to those reported in the above listed studies (e.g., Bock, 1986; Bock & Leobell, 1990) to

reuse the structure of the priming sentences in their responses.

Another finding for the priming analyses reported in the present study indicated that for

two of the three structures examined in the present study (i.e., transitive and two-clause

structures), participants in the two study groups did not show significant differences in the

priming effect. This result is not consistent with the only study that has examined the syntactic

priming effect in persons who stutter. Specifically, Anderson and Conture (2004) reported that

children who stutter seem to exhibit a great syntactic priming effect than children who do not

stutter. The authors argued that children who stutter might be "less skilled in morphosyntactic









construction processes" (p. 564) than children who do not stutter. This in turn might affect the

efficiency by which they express an intended message because these children might have fewer

computational resources available for syntactic processing. Within this view, children who

stutter might benefit from syntactic priming because they might take advantage of previously

activated forms (i.e., those of the priming sentences) to more effectively generate the intended

syntactic forms which in turn might reduce the time as observed in shorter SRTs for the primed

sentences in that group. As mentioned above, the results of the present study did not show a

difference in the extent of priming effect in the two groups. The differences between the results

reported in the present study and those reported by Anderson and Conture (2004) might be

attributed to one main reason. Namely, studies examining syntactic processing abilities in adults

who stutter have consistently shown that although syntax might play a role in stuttering during

the childhood years (e.g., Bernstein Ratner and Sih, 1987; Gaines, Runyan, & Meyers, 1991;

Gordon, Luper, & Peterson, 1981; Logan & Conture, 1995; 1997; Wall, Starkweather, & Cairns,

1981; Yaruss, 1999) it might not be the main contributing factor to stuttering during adolescence

and adulthood (e.g., Silverman and Ratner, 1997; Logan, 2001; 2003). The authors of the later

group of studies suggested that although some syntactic structures continue to be challenging for

both adolescents and adults who do not stutter, the effect of syntax in general and syntactic

complexity in specific seems to decrease as speakers grow older and their linguistic competency

improves. Within this view, it could be argued that children who stutter may exhibit subtle

speech production deficits and the presence of a prime in the context preceding the production of

given structures may provide a 'priming boost' and thus assist in overcoming either the retrieval

or encoding difficulties that might underlie speech production in those children. Adults who

stutter on the other hand have a higher linguistic competency than children who stutter as a result









of 'language maturation process', and thus priming (syntactic priming in this case) might not

play as prominent a role during speech processing as it would in children resulting in a smaller

gain of syntactic priming similar to the reported in the present study.

In terms of the effect of specific syntactic primes on frequency of syntactic structures in

responses, the results of the present study indicated that the structure of some syntactic primes

did affect the probability of the participant reusing the structure in those primes. The following

section presents the findings for the transitive, dative, and two-clause priming structures used in

the present study and discusses these results in comparison with earlier findings.

The first finding from the transitive priming experiment indicated that for all participants,

the structure of the transitive priming sentence did not seem to significantly affect the frequency

with which participants produced passive responses. Specifically, for all participants 18% of the

responses following passive primes were passive, 18% following active primes were passive, and

13% following locative primes were passive. In addition, and as can be observed from these

percentages, the frequency of passive responses following locative primes was lower than that

following both passive and active primes. These findings seem inconsistent with the ones

reported by Altmann et al. (2006) and Bock et al. (1990) who reported an overall significant

increase in passive responses after passive primes in both older and younger speakers. The

findings also seem inconsistent with those reported by Bock et al. (1990) who suggested that the

exposure to locative primes significantly increases the probability of reusing a passive prime in

the response. As the reader may recall, half the participants in the present study were college age

(7 out of 14 participants). Altmann et al. (2006) had explained that participants in Bock et al's

study might have produced an overall high proportion of passive responses and passive

responses following both passive and locative primes because those participants were college










age, and may have experienced frequent exposure to passive structures in lectures, thus making

those structures more familiar to those participants and resulting in a higher probability of using

those passive structures in their responses. Still, in the present study, the exposure to passive and

locative primes did not seem to significantly increase the probability of reusing a passive

structure in the response for the participants overall. It could be argued that the number of

participants in the Bock et al. study was much higher than that in the present study and thus

might have resulted in a more sensitive examination of the transitive primes in those studies. One

suggestion for future research would be to include many more participants and thus increase the

sensitivity of the study procedures to capture any effect the transitive primes may have on the

production of passives.

Different from the results of the transitive prime experiment, two significant Eindings were

reported in the dative prime experiment. The first Einding indicated that overall, participants

produced significantly more prepositional-obj ect responses following prepositional-obj ect

primes than following double-obj ect or obj ect-complement primes. The second result indicated

that participants in the two study groups exhibited a different pattern in terms of the structure of

the sentences they produced following the examined dative primes. Specifically, the PWS

produced significantly more prepositional-obj ect responses following double-obj ect primes

(43%) than following object-complement primes (16%). The PWNS on the other hand exhibited

the opposite pattern producing more prepositional-obj ect responses following obj ect-complement

primes (3 0%) than following double-obj ect primes (18%). The difference in the structure of the

dative response for the PWNS was not statistically significant. It is interesting to speculate what

factors might underlie the difference in the frequency of prepositional-responses between the two

groups. Recall that the dative primes which the participants repeated in the present study









exhibited three structures. The first dative prime was that of a double-obj ect dative verb as in

'The actress sold the stage manager her j ewelry.' The second dative prime was that of a

prepositional-obj ect dative verb such as in 'The actress sold her j ewelry to the stage manager.'

The third prime used in the dative experiment was that of an obj ect-complement verb as in 'The

stage manager nominated Mary best actress.' The sentences associated with the first two dative

verb forms (i.e., double-obj ect and prepositional-obj ect) share a common semantic representation

or theme in that both dative sentences include an agent (e.g., Mary), and two obj ects (e.g., boy

and paintbrush). The double-obj ect and obj ect-complement primes on the other hand share the

surface noun phrase structure, in that they both have a first noun phrase, followed by a verb,

which in turn is followed by two noun phrases. When participants are exposed to one of the

above described dative primes, and if priming is driven by the thematic aspects of the primes,

one would expect a comparable frequency of prepositional-obj ect responses following both

prepositional-obj ect primes and double-obj ect primes. However, if priming is driven by the

structural aspects of the primes, one would expect a comparable frequency of prepositional-

obj ect responses following both prepositional-obj ect and obj ect-complement primes. As reported

earlier in this chapter and in the Results chapter, the PWS produced most prepositional-obj ect

responses following prepositional-obj ect and double-obj ect primes, thus, it could be argued that

perhaps, priming of dative sentence forms in the PWS group is more driven by the semantic

aspects of the response rather than the structural aspect. On the other hand, the PWNS in the

present study produced more prepositional-obj ect responses following prepositional-object

primes and a comparable number following obj ect-complement primes. Based on the above, it

could be argued that perhaps, priming dative sentence forms in the PWNS group is more driven










by the structural aspects of the available options and in the PWS is more driven by the thematic

aspects of the available options.

The results of the two-clause prime experiment were similar to those of the transitive

prime experiment and provided one main finding. Specifically, and for all participants, the

structure of the two-clause priming sentence did not seem to significantly affect the frequency by

which participants produced embedded responses. Specifically, and although participants in the

two study groups exhibited a tendency to produce two-clause responses following two-clause

primes, the presence of the two-clause prime did not significantly increase the probability of

using the structure of that prime in the response. In terms of frequency of embedded responses

following the examined two-clause primes, 54% of the responses following conj oined primes

were embedded, 46% following center-embedded were embedded, and 51% following right-

embedded primes were embedded. This finding is consistent with the one reported by Altmann et

al. (2006) who reported that the structure of the "complex" (i.e., two-clause) priming sentence

did not seem significantly to affect the frequency by which older speakers produced embedded

responses.

Differences Between The Two Groups On the Prestudy Tasks

Recall that all participants in the present study Einished a number of language and memory

tasks as part of the participant screening procedures. The tasks were digits forward, digits

backward, digit ordering, and WAIS vocabulary test. The results of the statistical analyses

comparing the means of the two groups on those tasks indicated significant differences between

the two groups on two of those four tests. Namely, the mean scores for the PWNS were

significantly higher on the WAIS vocabulary and the digit ordering tests. Although differences

on these tests had the potential of confounding any potential differences between the groups on

of several of the analyses conducted in the present study, the responses of participants in the two










study groups were comparable on the maj ority of the examined measures and thus, did not seem

to be affected by the factors driving the differences on the tasks described above. Additionally, it

appears that whatever role memory would play in study tasks similar to the ones used in the

present study, and based on the fact that there were not significant differences between groups on

those study tasks, it is reasonable to suggest that the role memory plays in those tasks might not

be strong enough to significantly affect speech production. It is interesting to speculate what may

have driven those differences in scores on the prestudy language and memory tasks. One

suggestion for the difference in scores on those tests could be based on a common issue in

stuttering research. Namely, many of the persons who stutter who volunteer to participate in

stuttering research come from a population representing a broad range of socioeconomic and

educational background. The control group on the other hand might in many cases (one of which

is the present study) represent a narrower range of the population (i.e., college age range) and

thus reflect less diversity in that regard than the other group. Although most of the participants in

the present study were matched on as many factors as possible, still, it could be argued that the

PWS did represent a broader group while the PWNS represented a diverse group.

Conclusion

In the present study, the syntactic priming methodology was used to examine the effect of

syntactic structure on speech fluency and speech timing in adults who do and do not stutter. The

results suggest that although there were some differences between the study groups in processing

certain sentence types (i.e., datives) overall, syntactic processes involved in speech planning in

persons who stutter might be performed in a manner similar to those processes in persons who do

not stutter. This suggestion is based on the findings that speech initiation times and speech

fluency were similarly affected in participants in the two study groups by the structure of the

sentences produced by those participants. The present study is one of only few studies that have









used the syntactic priming methodology to examine speech production processes in adults who

stutter. The results support findings from a number of previous studies and show that syntactic

priming is a feasible methodology to study sentence processing abilities in persons who stutter.

In addition, and because participants tended to produce a variety of structures in their responses,

this methodology could assist in examining the effect of infrequent syntactic structures such as

passives and datives on speech production in persons who stutter.



















DESCRIPTION OF SPEECH AND LANGUAGE:

1. Have you ever had or currently have difficulty with:

la. Speech? Yes _No _. If "yes", please describe.


APPENDIX A
PRESTUDY TESTS

Background Survey: Control Group


Participant ID


Today's Date


lb. Oral or written language? Yes


.If "yes", please describe.


Ic. Fine or gross motor coordination? Yes_


2. Do you feel you hear normally? Yes _


3. Do you have difficulty remembering thing


4. Do you feel that you have normal vision? `

5. How would you describe your current heal
Excellent Good Fair Poor

6. Are you currently taking any medication?


.If "yes", please describe.


No _. If "no", please describe.


s? Yes No if "yes", please describe.


Yes No


.If "yes" please describe.


7. Does the medication affect your communication?



ACADEMIC EDUCATIONAL (skip (a) if not currently attending school)
1. Do you currently attend school? Yes _No _(if "no" skip to (b))
a. Name of school:

b. Year in school:

c. Major area of study:

d. Highest degree earned:











Background Survey: Experimental Group


Participant ID Today's Date


DESCRIPTION OF SPEECH AND LANGUAGE:

1. Have you ever had or currently have difficulty with speech? Yes No If yes,

please describe.






2. At what age was your problem first noticed?

3. Who noticed the problem?

4. How has the problem changed since that time?




5. Do other people notice your speech problem?

6. What do you believe caused the problem?

7. Have you previously had your speech, language, hearing skills evaluated?

Yes _No _. If "yes", please provide details.




8. Have you even had speech or language therapy? Yes _No _. If "yes":
Where :

With whom:

Dates:

Focus:

Re sults:

9. Are you currently applying any strategies that you may have learned during speech or

language therapy? Yes _No _. If "yes" please describe.










10. Have you ever had or currently have problems with written language?

Yes _No _. If "yes", please describe.


11. Have you ever had problems with fine or gross motor coordination?

Yes _No _. If "yes", please describe.


12. Do you feel you hear normally? Yes


.If "no", please describe.


13. Do you feel that you have normal vision? Yes _


14. Do you have difficulty remembering things? Yes


.If "no", please describe.


No _, if "yes", please describe.


15. Do any of your close family members have a history of speech, language, hearing, or

neurological problems? Yes _No _. If "yes" please explain.


16. How would you describe your current health?
Excellent Good Fair Poor

17. Are you currently taking any medication? Yes _


18. Does the medication affect your communication?


No _. If "yes" please describe.


ACADEMIC EDUCATIONAL
19. Do you currently attend school? Yes
e. Name of school:

f. Year in school:

g. Major area of study:
h. Highest degree earned:


No (if no kip to (b))










Pre-study Tests Used with Participants Who Stutter
Self-Ratinzg Scale

Participant ID# Today's Date

Please respond to each of the following statements by circling the point along the seven-
point scale that corresponds to your impressions of your speech. If a particular item does
not pertain to you, please write "not applicable" in the margin.

1. I participate in group-discussions with friends.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

2. I participate in group-discussions with family.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

3. I participate in group-discussions during class or at work.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

4. I stutter when talking with family members.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

5. I enjoy talking with others.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

6. My muscles feel tense when I speak.
1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

7. I feel out of breath when I am speaking.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never











8. Stuttering limits my academic and/or professional performance.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

9. I enjoy speaking before groups of people.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

10. I stutter when talking with friends.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

11. I feel embarrassed by the way I talk.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

12. I feel nervous when I am speaking to adults.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

13. Stuttering limits my ability to communicate with teachers and/or employees.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

14. Stuttering limits my ability to communicate with friends.

1 ----------2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

15. I repeat sounds or syllables when I am talking.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

16. If I expect to stutter on a certain word, I will substitute another word in its place.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never











17. I become "blocked" on speech sounds when I am talking.

1 ----------2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

18. I avoid speaking with others because of my stuttering.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

19. When talking with others, I acknowledge the fact that I stutter.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

20. My academic skills are better than average.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

21. I think that, someday, I will be able to manage my stuttering.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

22. I do whatever is necessary to hide my stuttering from others.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

23. I feel discouraged about the way I talk.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

24. Other people react negatively to my stuttering.

1 ---------- 2 ---------- 3 ---------- 4 ---------- 5 ---------- 6 ----------7

Always Sometimes Never

Comments :














I want you to talk for 3 minutes about a movie you saw.


















I want you to talk for 3 minutes about a pleasant experience.


Participant ID#


Date Severity Level


Stutterinzg Severity Task


Mean duration of three longest disfluencies:

Concomitants :

Number of disfluencies in 600 syllables: _

Suggested severity level:

Comments :


_; % of disfluencies in 100 syllables:









APPENDIX B
PRIMING AND FILLER SENTENCES AND PICTURES

A ctive/Passive/Locative

LIST A
Active
The students were bankrupted by the new sports complex.
The tourist was confused by the blinking traffic light.
The 747 was alerted by the airport's control tower.
The businessman was paged by the airline ticket agent.
The scientist was inspired by the apple tree.
The minister was cut by the broken stained glass window.

Passive
The cub scouts enjoyed the camp fire.
The construction worker drove the bulldozer.
The missing geologist underestimated the volcano.
The secretary cleaned the drinking fountain.
The quarterback destroyed the jukebox in the bar.
The surfer watched the stormy sea.

Locative
The stockbroker was sitting by the client.
The woman was swimming by the jellyfish.
The lumberj ack was eating by the giant redwood tree.
The dog was barking by the fence.
The councilman was strolling by the new building.
The ship was docking by the pier.

LIST B
Active
The stockbroker impressed the client.
The woman caught the jellyfish.
The lumberj ack struck the giant redwood tree.
The dog jumped the fence.
The councilman opened the new building.
The ship approached the pier.

Passive
The cub scouts were burned by the camp fire.
The construction worker was hit by the bulldozer.
The missing geologist was smothered by the volcano.
The secretary was splashed by the drinking fountain.
The quarterback was annoyed by the jukebox in the bar.
The surfer was excited by the stormy sea.









Locative
The students were working by the new sports complex.
The tourist was loitering by the blinking traffic light.
The 747 was landing by the airport's control tower.
The businessman was waiting by the airline ticket counter.
The scientist was sleeping by the apple tree.
The minister was praying by the broken stained glass.

LIST C
Active
The students tried the new sports complex.
The tourist misunderstood the blinking traffic light.
The 747 radioed the airport's control tower.
The businessman left the airline ticket counter.
The scientist examined the apple tree.
The minister fixed the broken stained glass window.

Passive
The stockbroker was sued by the client.
The woman was stung by the j ellyfish.
The lumberj ack was struck by the giant redwood tree.
The dog was protected by the fence.
The councilman was impressed by the new building.
The ship was damaged by the pier.

Locative
The cub scouts were singing by the camp fire.
The construction worker was digging by the bulldozer.
The missing geologist was wandering by the volcano.
The secretary was typing by the drinking fountain.
The quarterback was drinking by the jukebox in the bar.
The surfer was running by the stormy sea.

Double-Object Dative/Prepositional Dative/Object Complement

LIST A
Double-Object Dative
The mother read her older children a story
The volunteers sold the children some submarine sandwiches.
The candidate wrote the Republican mayor a letter.
The ambassador's secretary sent the marine a watch.
The freshman took the Salvation Army some clothes.
The breeder showed the young family his best dogs.










Prepositional Dative
The king promised his daughter to the triumphant knight.
The choir sang a new hymn to the wedding guests.
The defendant told a lie to the suspicious lawyer.
The diplomat took the secret documents to the President.
The president told a joke to a Russian reporter.
The widow rented the upstairs rooms to student tenants.

Object Complement
The students named the book a classic.
The theater manager nominated Mary Best Actress.
The housewife considered the church her second home.
That couple's son made their garage an apartment.
Proctor and Gamble appointed the drugstore owner chairman.
The children elected the nurses' group their favorite volunteers.

LIST B
Double-Object Dative
The king promised the triumphant knight his daughter.
The choir sang the wedding guests a new hymn.
The defendant told the suspicious lawyer a lie.
The diplomat took the President the secret documents.
The president told the Russian reporter a joke.
The widow rented the student tenants the upstairs rooms.

Prepositional Dative
The students brought a book to Stella.
The actress sold her j ewelry to the stage manager.
The housewife mailed a check to the bankrupt church.
The couple rented the garage to their neighbor's son.
Proctor and Gamble sent samples of the new detergent to consumers.
The children sang a song to the nurses' group.

Object Complement
The mother appointed the older children "official babysitters."
The volunteers considered submarine sandwiches the perfect lunch.
The Republican party elected the candidate mayor.
The ambassador's secretary assigned the marine Captain of the Watch.
The freshmen named the Salvation Army their favorite charity.
The breeder made family pets his top priority

LIST C
Double-Object Dative
The students brought Stella a book.
The actress sold the stage manager her j ewelry.
The housewife mailed the minister's bankrupt church a check.









The couple rented the neighbor's son their garage.
Proctor and Gamble sent consumers samples of the new detergent
The children sang the nurses' group a song.

Prepositional Dative
The mother read a story to her older children.
The volunteers sold some submarine sandwiches to the children.
The candidate wrote a letter to the Republican mayor.
The ambassador's secretary sent a watch to the marine.
The freshman took some clothes to the Salvation Army.
The breeder showed his best dogs to the young family.

Object Complement
The king's daughter considered the triumphant knight her fiance.
The choir considered the new hymn their favorite song.
The suspicious lawyer deemed the defendant a liar.
The diplomat made delivering the documents his primary mission.
The president appointed the Russian reporter his joke writer.
The widow deemed the upstairs room a disaster area.

Conzjoined/Center-Embedded/Right-Embedded

LIST A
Conjoined
The man spilled the wine and he stained the carpet.
The monkey looked under the bowl and then he hid his food.
The girl interrupted the boy who was talking very loudly.
The gambler chose the number and he won the lottery.
The mother punished the boy and then she broke the vase.
The girl smelled the flower and it reminded her of her grandmother.

Center-Embedded
The boys that teased the girl answered all the questions.
The man that wore white pants had stains on his knees.
The woman that scolded the boy climbed the tree.
The carpenter that hit the man was riding a bike.
The boy that was annoying the woman was talking on the phone.
The spy that carried the briefcase had the confidential documents.

Right-Embedded
The mechanic adjusted the fitting that closed the valve.
The carpenter drove the car that won the race.
The man kicked the dog that was chasing the cat.
The student bought a calculator that does calculus.
The woman wore a locket that was her grandmother's.
The waitress threw the dart that hit the bull's eye.









LIST B
Conjoined
The mechanic adjusted the fitting and then he closed the valve.
The driver drove the car and he won the race.
The man kicked the dog and then he chased the cat.
The student bought a calculator because she did calculus.
The woman wore a locket and it was her grandmother's.
The waitress threw the dart and she hit the bull's eye.

Center-Embedded
The man that spilled the wine stained the carpet.
The monkey that looked under the bowl was hiding his food.
The girl that interrupted the boy was humming very loudly.
The gambler that chose the number won the lottery.
The mother that punished the boy dropped the vase.
The girl that smelled the flower was very pretty.

Right-Embedded
The boys teased the girl who answered all the questions.
The man wore white pants that had stains on the knee.
The woman scolded the boy who had climbed the tree.
The carpenter hit a man who was riding a bike.
The boy was annoying the woman who was talking on the phone.
The spy carried the briefcase that had the confidential documents.

LIST C
Conjoined
The boys teased the girl because she answered all the questions.
The man wore white pants and they had stains on the knee.
The woman scolded the boy because he climbed the tree.
The carpenter hit a man, but he rode away on a bike.
The boy was annoying the woman because she was talking on the phone.
The spy carried the briefcase because it held the confidential documents.

Center-Embedded
The mechanic that adjusted the fitting closed the valve.
The driver that drove the car won the race.
The man that kicked the dog was chasing the cat.
The student that bought a calculator takes calculus.
The woman that wore a locket was a grandmother.
The waitress that threw the dart hit the bull's eye.

Right-Embedded
The man spilled the wine that stained the carpet.
The monkey looked under the bowl that was hiding his food.
The girl interrupted the boy that was humming very loudly.










The gambler chose the number that won the lottery.
The mother punished the boy that dropped the vase.
The girl smelled the flower that was very pretty.

Filler Sentences

The books were expensive
The detective noticed the cuts on the bartender's hand
It was difficult to forget the photographs of the concentration camp
The acorns fell from the top of the oak tree.
The boy seems to enj oy the ice cream.
Nancy wants to redecorate the family room.
It was easy to hide the money.
The house took fifty years to build
Henry Vlll married Jane Seymour
Bob Dole used to be in the Senate.
The desk is in the room.
The surgeon cut himself with a scalpel.
The staff burned the incriminating papers.
Jackhammers are noisy.
The stockbroker was angry about the rising interest rates.
The freezing rain made the street slippery.
There is a red spot on Jupiter.
There were thirteen original colonies.
The stuntman threw himself out of the window.
It was tough to fail the tennis class.
That Billy was unhappy was apparent from his expression.
The president of UF is Bernie Machen.
George Bush is the president.
There were several witnesses to the three car accident.
The shaken victim described her assailant.
The freezing rain made the street slippery.
There were a lot of earthquakes last year.
The anger of the crowd was hard to understand.
There are thirty players in the tournament.
McDonald's is the largest restaurant chain in the world.
The weather was nice.
That eighteen-year old fulfilled all the requirements for a bachelor's degree.
The accident was inexplicable.
Working on a computer can be frightening.
The football fans lined up at the gate.
The refrigerator hasn't been defrosted in months.
The stockbroker was angry about the rising interest rates.
Liberace died after a serious illness.
The blouse had a button missing.
It was impossible to answer all the questions.









Shirley MacLaine played herself in one of her movies.
Some of the trees on University Ave. are diseased
A famous scientist is appearing on the Discovery Channel.
Independence Hall is on Chestnut Street in Philadelphia.
Robert E. Lee surrendered in Virginia
The calculator is broken.

Priming Sentence Sets and Matched Target Pictures
Transitive Sentences
Priming sentence set:
Active: The students tried the new sports complex.
Passive: The students were bankrupted by the new sports complex.
Locative: The students were working by the new sports complex.
Target picture: Toy startles little girl

Priming sentence set:
Active: The cub scouts enjoyed the camp fire.
Passive: The cub scouts were burned by the camp fire.
Locative: The cub scouts were singing by the camp fire.
Target picture: Ambulance hits policeman

Priming sentence set:
Active: The construction worker drove the bulldozer.
Passive: The construction worker was hit by the bulldozer.
Locative: The construction worker was digging by the bulldozer.
Target picture: Sailor kicking soldier

Priming sentence set:
Active: The missing geologist underestimated the volcano.
Passive: The missing geologist was smothered by the volcano.
Locative: The missing geologist was wandering by the volcano.
Target picture: Avalanche scares skiers

Priming sentence set:
Active: The secretary cleaned the drinking fountain.
Passive: The secretary was splashed by the drinking fountain.
Locative: The secretary was typing by the drinking fountain.
Target picture: Train runs into bus

Priming sentence set:
Active: The quarterback destroyed the jukebox in the bar.
Passive: The quarterback was annoyed by the jukebox in the bar.
Locative: The quarterback was drinking by the jukebox in the bar.
Target picture: Girl kicking a boy









Priming sentence set:
Active: The surfer watched the stormy sea.
Passive: The surfer was excited by the stormy sea.
Locative: The surfer was running by the stormy sea.
Target picture: Lightning strikes church


Priming sentence set:
Active: The stockbroker impressed the client.
Passive: The stockbroker was sued by the client.
Locative: The stockbroker was sitting by the client.
Target picture: Dog chases mailman

Priming sentence set:
Active: The woman caught the jellyfish.
Passive: The woman was stung by the jellyfish.
Locative: The woman was swimming by the jellyfish.
Target picture: Horse kicks cow

Priming sentence set:
Active: The tourist misunderstood the blinking traffic light.
Passive: The tourist was confused by the blinking traffic light.
Locative: The tourist was loitering by the blinking traffic light.
Target picture: Hydrant squirts fireman

Priming sentence set:
Active: The 747 radioed the airport's control tower.
Passive: The 747 was alerted by the airport's control tower.
Locative: The 747 was landing by the airport's control tower.
Target picture: Turtle squirts a mouse

Priming sentence set:
Active: The businessman left the airline ticket counter.
Passive: The businessman was paged by the airline ticket agent.
Locative: The businessman was waiting by the airline ticket counter.
Target picture: Wrecking ball destroys building

Priming sentence set:
Active: The minister fixed the broken stained glass window.
Passive: The minister was cut by the broken stained glass window.
Locative: The minister was praying by the broken stained glass.
Target picture: Tornado destroys barn

Priming sentence set:
Active: The scientist examined the apple tree.
Passive: The scientist was inspired by the apple tree.









Locative: The scientist was sleeping by the apple tree.
Target picture: Woman kisses man

Priming sentence set:
Active: The councilman opened the new building.
Passive: The councilman was impressed by the new building.
Locative: The councilman was strolling by the new building.
Target picture: Rock hits boy on head

Priming sentence set:
Active: The ship approached the pier.
Passive: The ship was damaged by the pier.
Locative: The ship was docking by the pier.
Target picture: Boy is saving a girl from drowning

Priming sentence set:
Active: The lumberj ack struck the giant redwood tree.
Passive: The lumberj ack was struck by the giant redwood tree.
Locative: The lumberj ack was eating by the giant redwood tree.
Target picture: Fireman rescues baby from fire

Priming sentence set:
Active: The dog jumped the fence.
Passive: The dog was protected by the fence.
Locative: The dog was barking by the fence.
Target picture: Bee stings man

Dative Sentences

Priming sentence set:
Double-Obj ect Dative: The candidate wrote the Republican mayor a letter.
Prepositional Dative: The candidate wrote a letter to the Republican mayor.
Obj ect Complement: The Republican Party elected the candidate mayor.
Target picture: Waitress give menu to man

Priming sentence set:
Double-Obj ect Dative: The mother read her older children a story.
Prepositional Dative: The mother read a story to her older children.
Obj ect Complement: The mother appointed the older children "official babysitters."
Target picture: Boy giving apple to teacher

Priming sentence set:
Double-Obj ect Dative: The volunteers sold the children some submarine sandwiches.
Prepositional Dative: The volunteers sold some submarine sandwiches to the children.
Obj ect Complement: The volunteers considered submarine sandwiches the perfect lunch.
Target picture: Guy gives guitar to musician











Priming sentence set:
Double-Obj ect Dative: The ambassador's secretary sent the marine a watch.
Prepositional Dative: The ambassador's secretary sent a watch to the marine.
Obj ect Complement: The ambassador's secretary assigned the marine Captain of the Watch.
Target picture: Girl hands paintbrush to boy on ladder

Priming sentence set:
Double-Obj ect Dative: The freshmen took the Salvation Army some clothes.
Prepositional Dative: The freshmen named the Salvation Army their favorite charity.
Obj ect Complement: The freshmen named the Salvation Army their "favorite charity."
Target picture: Salesman shows car to a couple

Priming sentence set:
Double-Obj ect Dative: The breeder showed the young family his best dogs.
Prepositional Dative: The breeder showed his best dogs to the young family.
Obj ect Complement: The breeder made family pets his top priority.
Target picture: Boy and girl give flowers to the man

Priming sentence set:
Double-Obj ect Dative: The king promised the triumphant knight his daughter.
Prepositional Dative: The king promised his daughter to the triumphant knight.
Obj ect Complement: The king's daughter considered the triumphant knight her flance.
Target picture: Cowboy gives hat to clown

Priming sentence set:
Double-Obj ect Dative: The choir sang the wedding guests a new hymn.
Prepositional Dative: The choir sang a new hymn to the wedding guests.
Obj ect Complement: The choir considered the new hymn their favorite song.
Target picture: Girl reads to boy

Priming sentence set:
Double-Obj ect Dative: The defendant told the suspicious lawyer a lie.
Prepositional Dative: The defendant told a lie to the suspicious lawyer.
Obj ect Complement: The suspicious lawyer deemed the defendant a liar.
Target picture: Waitress served drinks to man

Priming sentence set:
Double-Obj ect Dative: The diplomat took the President the secret documents.
Prepositional Dative: The diplomat took the secret documents to the President.
Obj ect Complement: The diplomat made delivering the documents his primary mission.
Target picture: Boy and girl show picture to the teacher

Priming sentence set:
Double-Obj ect Dative: The president told the Russian reporter a j oke.
Prepositional Dative: The president told a joke to a Russian reporter.










Obj ect Complement: The president appointed the Russian reporter his j oke writer.
Target picture: Lawyer shows gun to judge

Priming sentence set:
Double-Obj ect Dative: The widow rented the student tenants the upstairs rooms.
Prepositional Dative: The widow rented the upstairs rooms to student tenants.
------: The widow deemed the upstairs room a disaster area.
Target picture: Man is passing pitcher to woman

Priming sentence set:
Double-Obj ect Dative: The students brought Stella a book.
Prepositional Dative: The students brought a book to Stella.
Obj ect Complement: The students named the book a classic.
Target picture: Nurse gives stethoscope to doctor

Priming sentence set:
Double-Obj ect Dative: The actress sold the stage manager her j ewelry.
Prepositional Dative: The actress sold her j ewelry to the stage manager.
Obj ect Complement: The stage manager nominated Mary Best Actress.
Target picture: Boy gives valentine to girl

Priming sentence set:
Double-Obj ect Dative: The housewife mailed the minister's bankrupt church a check.
Prepositional Dative: The housewife mailed a check to the bankrupt church.
Obj ect Complement: The housewife considered the church her second home.
Target picture: Woman throws stick to dog

Priming sentence set:
Double-Obj ect Dative: The couple rented the neighbor's son their garage.
Prepositional Dative: The couple rented the garage to their neighbor's son.
Obj ect Complement: That couple's son made their garage an apartment.
Target picture: Cop gives ticket to man

Priming sentence set:
Double-Obj ect Dative: Proctor and Gamble sent consumers samples of the new detergent.
Prepositional Dative: Proctor and Gamble sent samples of the new detergent to consumers.
Obj ect Complement: Proctor and Gamble appointed the drugstore owner chairman.
Target picture: Librarian gives book to boy

Priming sentence set:
Double-Obj ect Dative: The children sang the nurses' group a song.
Prepositional Dative: The children sang a song to the nurses' group.
Obj ect Complement: The children elected the nurses' group their favorite volunteers.
Target picture: Waitress gives menu to man









Conjoined-Embedded Sentences


Priming sentence set:
Conj oined: The man spilled the wine and he stained the carpet.
Center-Embedded: The man that spilled the wine stained the carpet.
Right-Embedded: The man spilled the wine that stained the carpet.
Target picture: Boy chasing cat, cat chasing duck

Priming sentence set:
Conj oined: The monkey looked under the bowl and then he hid his food.
Center-Embedded: The monkey that looked under the bowl was hiding his food.
Right-Embedded: The monkey looked under the bowl that was hiding his food.
Target picture: Boy touching girl, girl touching monkey

Priming sentence set:
Conj oined: girl interrupted the boy that was humming very loudly.
Center-Embedded: The girl that interrupted the boy was humming very loudly.
Right-Embedded: The girl interrupted the boy who was talking very loudly.
Target picture: Cat and boy kicking girl

Priming sentence set:
Conj oined: The gambler chose the number and he won the lottery.
Center-Embedded: The gambler that chose the number won the lottery.
Right-Embedded: The gambler chose the number that won the lottery.
Target picture: Dog chasing girl, girl chasing boy

Priming sentence set:
Conj oined: The mother punished the boy and then she broke the vase.
Center-Embedded: The mother that punished the boy dropped the vase.
Right-Embedded: The mother punished the boy that dropped the vase.
Target picture: Dog biting cat, cat biting girl

Priming sentence set:
Conj oined: The girl smelled the flower and it reminded her of her grandmother.
Center-Embedded: The girl that smelled the flower was very pretty.
Right-Embedded: The girl smelled the flower that was very pretty.
Target picture: Man pulling woman, woman pulling dog

Priming sentence set:
Conj oined: The mechanic adjusted the fitting and then he closed the valve.
Center-Embedded: The mechanic that adjusted the fitting closed the valve.
Right-Embedded: The mechanic adjusted the fitting that closed the valve.
Target picture: Bear leading girl, girl leading boy

Priming sentence set:
Conj oined: The driver drove the car and he won the race.









Center-Embedded: The driver that drove the car won the race.
Right-Embedded: The driver drove the car that won the race.
Target picture: Monkey patting girl, cat jumping on girl

Pricing sentence set:
Conj oined: The man kicked the dog and then he chased the cat.
Center-Embedded: The man that kicked the dog was chasing the cat.
Right-Embedded: The man kicked the dog that was chasing the cat.
Target picture: Boy watching bear spying on bird

Prining sentence set:
Conj oined: The student bought a calculator because she did calculus.
Center-Embedded: The student that bought a calculator takes calculus.
Right-Embedded: The student bought a calculator that does calculus.
Target picture: Girl pulling cat and chasing baby

Pricing sentence set:
Conj oined: The woman wore a locket and it was her grandmother's.
Center-Embedded: The woman that wore a locket was a grandmother.
Right-Embedded: The woman wore a locket that was her grandmother's.
Target picture: Girl pulling baby and dog

Pricing sentence set:
Conj oined: The waitress threw the dart and she hit the bull's eye.
Center-Embedded: The waitress that threw the dart hit the bull's eye.
Right-Embedded: The waitress threw the dart that hit the bull's eye.
Target picture: Bear spying on man looking at boy

Pricing sentence set:
Conj oined: The man wore white pants and they had stains on the knee.
Center-Embedded: The man that wore white pants had stains on his knees.
Right-Embedded: The man wore white pants that had stains on the knee.
Target picture: Cat kicking girl, girl kicking boy

Pricing sentence set:
Conj oined: The boys teased the girl because she answered all the questions.
Center-Embedded: The boys that teased the girl answered all the questions.
Right-Embedded: The boys teased the girl who answered all the questions.
Target picture: Boy trying to pat cat, cat scratching dog

Pricing sentence set:
Conj oined: The woman scolded the boy because he climbed the tree.
Center-Embedded: The woman that scolded the boy climbed the tree.
Right-Embedded: The woman scolded the boy who had climbed the tree.
Target picture: Woman staring at man, man watching baby









Priming sentence set:
Conj oined: The carpenter hit a man, but he rode away on a bike.
Center-Embedded: The carpenter that hit the man was riding a bike.
Right-Embedded: The carpenter hit a man who was riding a bike.
Target picture: Monkey patting boy, boy patting girl

Priming sentence set:
Conj oined: The boy was annoying the woman because she was talking on the phone.
Center-Embedded: The boy that was annoying the woman was talking on the phone.
Right-Embedded: The boy was annoying the woman who was talking on the phone.
Target picture: Woman pulling man and dog

Priming sentence set:
Conj oined: The spy carried the briefcase because it held the confidential documents.
Center-Embedded: The spy that carried the briefcase had the confidential documents.
Right-Embedded: The spy carried the briefcase that had the confidential documents.
Target picture: Girl leading bear, bear leading boy










LIST OF REFERENCES


Alvarez, S., Yimovines, B., Key-DeLyria, S., & Altmann, L. J. P. (November, 2006). Age
differences in the stimulability of sentence structure: Implications for treatment. Paper
presented at the annual meeting of the Ameri can Sp eech-Language-Heari ng As soci ati on,
Miami, FL.

Anderson, J., & Conture, E. (2004). Sentence-structure priming in young children who do and do
not stutter. Journal of Speech, Language and Hearing Research, 47, 552-571.

Bernstein Ratner, N., & Sih, C. C. (1987). Effects of gradual increases in sentence length and
complexity on children's disfluency. Journal of Speech and Hearing Disorders, 52, 278-
287.

Besche, C., Passerieux, C., Segui, J., Sarfati, Y., Laurent, J. P., & Hardy-Bayle, M. C. (1997).
Syntactic and semantic processing in schizophrenic patients evaluated by lexical-decision
tasks. Neuropsychology, 11(4):498-505.

Bloodstein, O. (1995). A handbook on stuttering (5th ed.). San Diego, CA: Singular.

Bock, K. (1986). Syntactic persistence in language production. Cognitive Psychology, 18, 355-
387.

Bock, K., & Loebell, H. (1990). Framing sentences. Cognition, 35, 1-39.

Bosshardt, H. G. (1993). Differences between stutterers' and nonstutterers' short-term recall and
recognition performance. Journal of Speech and Hearing Research, 36, 286-293.

Bosshardt, H. G., Ballmer, W., & de Nil, L. F. (2002). Effects of category and rhyme decisions
on sentence production. Journal of Speech, Language, and Hearing Research, 45, 844-
657.

Bosshardt, H. G., & Fransen, H. (1996). Online sentence processing in adults who stutter and
adults who do not stutter. Journal of Speech and Hearing Research, 39, 785-797.

Branigan, H. P., Pickering, M. J., McLean, J. F., & Stewart, A. (2006). The role of local and
global syntactic structure in language production: evidence from syntactic priming.
Language and Cognitive Processes, 21, 974-1010.

Burger, R., & Wijnen, F. (1999). Phonological encoding and word stress in stuttering and
nonstuttering subj ects. Journal of Fluency Disorders, 24, 91-106.

Caruso, A. J., Chozko-Zajko, W. J., Bidinger, D. A., & Sommers, R. K. (1994). Adults who
stutter: Response to cognitive stress. Journal of Speech and Hearing Research, 37, 738-
745.

Conture, E. G. (2001). Stuttering: Its nature, diagnosis, and treatment. Needham Heights, MA:
Allyn & Bacon.










Cuadrado, E. M., & Weber-Fox, C. (2003). Atypical syntactic processing in individuals who
stutter: Evidence from event-related brain potentials and behavioral measures. Journal of
Speech, Language, andHearingResearch,, 46, 960-976.

de Roo, E., Kolk, H., & Hofstede, B. (2003). Structural properties of syntactically reduced
speech: a comparison of normal speakers and Broca's aphasics. Brain and Language,
86(1):99-115.

Faust, M., Silber, A., & Kaniel, S. (2001). Evidence from sentence priming for an atypical
language organization in the brain of dyslexic males. Laterality, 6(1):39-56.

Ferreira, F. (1991). Effects of length and syntactic complexity on initiation times for prepared
utterances. Journal of~entory and Language, 30, 210-233 .

Gaines, N. D., Runyan, C. M., & Meyers, S. C. (1991).A comparison of stutterers' fluent versus
stuttered utterances on measures of length and complexity. Journal of Speech andHearing
Research, 34, 37-42.

Gibson, E. (1998). Linguistic complexity: locality of syntactic dependencies. Cognition, 68, 1-
76.

Gordon, P. A., Luper, H. L., & Peterson, H. A. (1986). The effects of syntactic complexity on the
occurrence of disfluencies in 5 year old nonstutterers. Journal of Fluency Disorders, 11,
151-164.

Greenbaum, S., & Quirk, R. (1996). A student 's grananar of the English language (10th ed.).
Essex: Longman.

Hartfield, K., & Conture, E. (2006). Effects of perceptual and conceptual similarity in lexical
priming of young children who stutter. Preliminary findings. Journal of Fluency
Disorders, 31, 303, 324.

Hartsuiker, R. J., & Kolk, H. H. (1998). Syntactic facilitation in agrammatic sentence production.
Brain and Language, 62, 221-254.

Jarvis, B. G. (2006). DirectRT (Version 2006 2.1~6) [Computer Software]. New York, NY:
Empirisoft Corporation.

Karniol, R. (1995). Stuttering, language, and cognition: A review and model of stuttering as
suprasegmental sentence plan alignment (SPA). Psycholinguistic Bulletin, 117, 1, 104-124.

Kolk, H., & Postma, A. (1997). Stuttering as a covert repair phenomenon. In R. Curlee & G.
Siegel (Eds.), Nature and treatment ofstuttering: New directions (2nd ed., pp. 182- 203).
Boston: Allyn & Bacon.

Levelt, W., J., M., & Kelter, S. (1982). Surface form and memory in question answering.
Cognitive Psychology, (14) 1, 78-106.










Logan, K. J. (2001). The effect of syntactic complexity upon the speech fluency of adolescents
and adults who stutter. Journal of Fhtency Disorders, 26, 85-106.

Logan, K. J. (2003). The effect of syntactic structure upon speech initiation times of stuttering
and nonstuttering speakers. Journal of Fluency Disorders, 28, 17-35.

Logan, K. J., & Conture, E. G. (1995). Length, grammatical complexity, and rate differences in
stuttered and fluent conversational utterances of children who stutter. Journal of Fluency
Disorders, 20, 35-62.

Logan, K. J., & Conture, E. G. (1997). Temporal, grammatical, and phonological characteristics
of conversational utterances produced by children who stutter. Journal of Speech and
Hearing Research, 40, 107-210.

Manning, W. H. (2000). Appeal of the demands and capacities model: Conclusions. Journal of
Fhtency Disorders, 25, 4, 377-383.

Melnick, K. S., Conture, E. G., & Ohde, R. N. (2003). Phonological priming in picture naming of
young children who stutter. Journal of Speech, Language, andHearingResearch,, 46,
1428-1443.

Nippold, M. A. (1990). Concomitant speech and language disorders in stuttering children: A
critique of the literature. Journal of Speech and Hearing Disorders, 55, 5 1-60.

Pellowski, M. W., & Conture, E. G. (2005). Lexical encoding in young children who do and do
not stutter. Journal ofSpeech, Language, andHearingResearch,, 48, 278-294.

Perkins, W. H., Kent, R. D., & Curlee, R. F. (1991). A theory of neuropsycholinguistic function
in stuttering. Journal of Speech andHearingResearch,, 34, 734-752.

Peters, T. J., & Guitar, B. (1991). Stuttering: An integrated approach to its nature and treatment.
Baltimore: Williams and Wilkins.

Pickering, M. J., & Branigan, H. P. (1999). Syntactic priming in language production. Trends in
Cognitive Sciences, 3, 136-141.

Postma, A., & Kolk, H. (1993). The covert repair hypothesis: Prearticulatory repair processes in
normal and stuttered disfluencies. Journal of Speech and Hearing Research, 36, 472-487.

Ratner, N. B. (1997). Stuttering: A psycholinguistic perspective. In R. Curlee & G. Siegel (Eds.),
Nature and treatment ofstuttering: New directions (2nd ed., pp. 99- 127). Boston: Allyn
& Bacon.

Riley, G. D. (1994). Stuttering Severity Instrument for Children and Adults--3. Austin, TX: Pro-
Ed.

Siegel, G. M. (2000). Demands and capacities or demands and performance? Journal of Fluency
Disorders, 25, 4,321-327.










Silverman, S., & Ratner, N. B. (1997). Syntactic complexity, fluency, and accuracy of sentence
imitation in adolescents. Journal of Speech, Language, and Hearing Research, 40, 95-106.

Smith, A., & Kelly, E. (1997). Stuttering: A dynamic, multifactorial model. In R. F. Curlee & G.
M. Siegel (Eds.), Nature and treatment ofstuttering: New directions (2nd ed., pp. 204-
217). Needham Heights, MA: Allyn & Bacon.

Smith, M., & Wheeldon, L. (2001). Syntactic priming in spoken sentence production--An online
study. Cognition, 78, 123-164.

Starkweather, C. W. (1987). Fluency and stuttering. Englewood Cliffs, NJ: Prentice-Hall.

Wall, M., Starkweather, C. W., & Cairns, H. S. (1981). Syntactic influences on stuttering in
young child stutterers. Journal of Fhtency Disorders, 6, 283-298.

Watkins R. V., Yairi E., & Ambrose, N. G. (1999). Early childhood stuttering III: initial status of
expressive language abilities. Journal ofSpeech, Language, and Hearing Research,
42(5):1125-35.

Weber-Fox, C., Spencer, R., Spruill, J., & Smith, A. (2004). Phonologic processing in adults who
stutter: Electrophysiological & behavioral evidence. Journal Speech Language and
Hearing Research, 47, 1244-1258.

Wijnen, F., & Boers, I. (1994). Phonological priming effects in stutterers. Journal of Fluency
Disorders, 19, 1-20.

Yaruss, J. S. (1999). Utterance length, syntactic complexity, and childhood stuttering. Journal of
Speech andHearingResearch,, 42, 329-344.

Yaruss, J. S. (2000). The role of performance in the demands and capacities model. Journal of
Fhtency Disorders, 25, 4, 347-358.









BIOGRAPHICAL SKETCH

Maisa Haj-Tas obtained a BA in English language and literature and an MA is speech

pathology in her home country Jordan. She was a teacher of English as a second language for

several years before she came to UF in 2001 to pursue a doctoral degree is speech pathology and

was a teaching assistant there until the end of fall, 2006. She worked on several research proj ects

with her advisor Dr. Kenneth Logan and presented several original papers at one international

and several national conventions.

Since early 2007 until the end of the Summer semester of that year, Maisa was a teaching

assistant at the English Language Institute at UF where she taught reading and writing skills to

intermediate level second language learners.

During her years at UF, Maisa was an active member in the International Student Speakers

Bureau and presented several talks to students at schools in Gainesville about her country and

culture. She was awarded the Alec Courtelis award for outstanding international student in 2006

for her contribution and activities at UF. She also received several other prestigious awards

among which were the Grinter Fellowship, the Dissertation Fellowship, in addition to several

travel awards from her department, college, and the graduate school at UF.

Upon completion of her doctoral work at UF, Maisa plans to go back to Jordan and will

j oin the faculty at the University of Jordan. She is proud to say that she is the first person in her

country of about five million people with a doctoral degree in stuttering. He future plans include

teaching, conducting research, and assisting in the fluency clinic.





PAGE 1

1 EFFECT OF SYNTACTIC STRUCTURE ON SPEECH PRODUCTION IN ADULTS WHO STUTTER By MAISA A. HAJ-TAS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Maisa A. Haj Tas

PAGE 3

3 To my parents: Atef Haj-Tas and Samiya Hakouz.

PAGE 4

4 ACKNOWLEDGMENTS My thanks and gratitude go to my adviso r Dr. Kenneth Logan for his kind assistance during my years at UF. I had the honor and pr ivilege of working with Dr. Logan on several research projects and clinic activities during the past six years. I learned a lot from Dr. Logan about how to do research, how to be a good c linician, and how to ex cel in teaching. Most importantly, and through his model, I learned abou t the qualities and value of a great mentor. I hope that one day, I could be of similar value to my students. I also want to thank my supervisory comm ittee Dr. Lori Altmann, Dr. Linda Lombardino, and Dr. Ratree Wayland for their feedback a nd support during the past years. Both Dr. Lombardino and Dr. Wayland provid ed valuable feedback on my work and were very supportive of my efforts. Dr. Altmann was especially helpfu l in assisting me to understand the details of the methodology I used in my study. She always found time to meet with me and was very generous in sharing and providing invaluable suggestions about how to better und erstand and analyze my data. I want to thank my friends bot h in the States and in Jordan for being there to share the great times and for their support during the rough times. It was always heart warming to know that although I was many miles aw ay from my family, I had many br others and sisters here to be with whenever I needed someone to talk to or someone to listen. I want to extend special thanks to my aunt Suad Bitamour, her husband Ibrahim Bitamour and sons Moaied and Mahmoud for their hos pitality and never ending support and encouragement. Staying with them a few weeks every year energized my efforts and reminded me of how wonderful it is to have family around. Most importantly, I thank my parents Samiya Hakouz and Atef Haj-Tas and my sisters Ghaida, Samar, Rasha, Reem, and Farah, and brothers Abdullah and Muhammad for always

PAGE 5

5 being there for me. Their continuous encouragem ent kept me going. My mother was my biggest supporter throughout this journey. She always show ed genuine interest in whatever I was doing, no matter how trivial that was, and always no matter how often or how late or early I called, found the time to listen. Without her support, I would not have been able to achieve much. I also want to thank the Department of Communication Sciences and Disorders and the English Language Institute for supporting my st udy during the past years. Through their support (both financial and moral), I was able to take classes and focus on improving my abilities as both a scholar and a teacher. My thanks also go to Janet Skotko at the Voice Institute and the clinicians at the Jacksonville Speech and Hearing Ce nter for facilitating meetings with participants for my study. Their assistance was invaluable and was a huge f actor in helping me fi nishing my dissertation. Finally, I want to thank everybody who partic ipated in my study. Although the study tasks took on average about 80 minutes to finish, all pa rticipants were gracious patient, and did their best to make my efforts a success.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......10 ABSTRACT....................................................................................................................... ............11 CHAPTER 1 INTRODUCTION..................................................................................................................13 Psycholinguistic Models of Stuttering: An overview.............................................................14 The Covert Repair Hypothesis (CRH)............................................................................14 The Neuropsycholinguistic Theory of Stuttering............................................................15 The Suprasegmental Alignment Model of Stuttering......................................................16 The Demands and Capacities Model of Stuttering (DCM).............................................16 The Multifactorial Model of Stuttering...........................................................................17 Linguistic Factors and Stuttering............................................................................................19 Methodological Limitations....................................................................................................22 The Syntactic Priming Methodology......................................................................................24 Syntactic Priming and Speakers with Typical Language Production.............................25 Syntactic Priming and Speakers wi th Atypical Language Production............................29 The Priming Methodology and People who Stutter...............................................................30 Phonological Priming and PWS......................................................................................31 Lexical/Semantic Priming and PWS...............................................................................34 Syntactic Priming and PWS............................................................................................36 Summary........................................................................................................................ .........37 Research Questions............................................................................................................. ....38 Question One...................................................................................................................39 Question Two..................................................................................................................39 Question Three................................................................................................................39 Question Four..................................................................................................................39 2 METHODS........................................................................................................................ .....40 Participants................................................................................................................... ..........40 Inclusion and Exclusion Criteria............................................................................................40 Material....................................................................................................................... ............41 Pre-Study Material Used with all Participants................................................................41 Background survey (Appendix A)...........................................................................41 WAIS vocabulary.....................................................................................................42 Digits forward and digits backward.........................................................................42 Digit ordering...........................................................................................................43

PAGE 7

7 Pre-Study Material Used with PWS................................................................................43 Stuttering severity leve l task (Appendix A).............................................................43 Self-rating scale (Appendix A)................................................................................44 Preparation and Description of the Mate rial Used in Syntactic Priming Task................44 Transitive priming sentences and pictures...............................................................44 Dative priming sentences and pictures.....................................................................46 Two-clause priming sentences and pictures.............................................................47 Filler sentences and pictures....................................................................................48 Apparatus...................................................................................................................... ..........49 Procedures..................................................................................................................... ..........50 Data Collection................................................................................................................50 Data Preparation for Analyses.........................................................................................51 Data excluded from the final analyses.....................................................................52 General descriptive analyses....................................................................................54 Data preparation for speech fluency analyses..........................................................55 Data preparation for sentence type analyses............................................................56 Data preparation for SRT analyses...........................................................................58 Intrajudge and Interjudge Measurement Reliability........................................................60 3 RESULTS........................................................................................................................ .......65 Fluency-Related Results........................................................................................................ .65 Transitive Experiment.....................................................................................................66 Dative Experiment...........................................................................................................68 Two-Clause Experiment..................................................................................................70 Fluency Across Picture Types.........................................................................................72 Priming Analyses............................................................................................................... .....75 Transitive..................................................................................................................... ....76 Dative......................................................................................................................... .....76 Two-Clause..................................................................................................................... .80 Summary of the Priming Analyses..................................................................................81 SRT Analyses................................................................................................................... ......81 4 DISCUSSION..................................................................................................................... ....92 The Effect of Syntactic Structure on Fluency.........................................................................93 The Effect of Picture Type on Fluency...................................................................................98 The Effect of Syntactic Stru cture on Speech Reaction Time.................................................99 The Effect of Syntactic Priming on Response Type.............................................................105 Differences Between The Two Groups On the Prestudy Tasks...........................................110 Conclusion..................................................................................................................... .......111 APPENDIX A PRESTUDY TESTS.............................................................................................................113 B PRIMING AND FILLER SENTENCES AND PICTURES................................................120

PAGE 8

8 LIST OF REFERENCES.............................................................................................................134 BIOGRAPHICAL SKETCH.......................................................................................................138

PAGE 9

9 LIST OF TABLES Table page 2-1 Participant Demographics and Performa nce on the Pre-study Language and Memory Tests.......................................................................................................................... .........64

PAGE 10

10 LIST OF FIGURES Figure page 2-1 Mean and standard deviation for th e number of responses produced across the examined sentence types (Transitive, Da tive, and Two-Clause) in addition to responses that exhibited synt actic structures other than the examined types (Other)........62 2-2 Mean and standard deviation for number of words produced for (a) Transitive responses in the Transitive experiment (b) Dative responses in the Dative experiment, and (c) Two-Clause respons es in the Two-Clause experiment......................62 2-3 Mean and standard deviation for the percentage of sentences produced when the sentence was entirely fluent, when the fi rst noun phrase in the sentence was fluent, and when the first word in the first noun phrase in the sentence was fluent.....................63 3-1 Mean number and standard error for repetitions and prolong ations (Rep.& Pro.) and other types of disflu encies (Other) when the priming sentences were (a) Transitive, (b) Dative, and (c) Two-Clause.......................................................................86 3-2 Mean number and standard error for repetitions and prolong ations (Rep.& Pro.) and other types of disfluenci es (Other) in responses when the priming pictures were (a) Transitive (b) Dative, and (c) Two-Clause...................................................................88 3-3 Mean percentage and standard error fo r responses when the priming sentences were (a) Transitive, (b) Dative, and (c) Two-Clause..................................................................89 3-4 Mean and standard error for speech r eaction time (SRT) when the responses were(a) Transitive in the Transitive experiment, (b) Dative in the Dative experiment, and (c) Two-Clause in the Two-Clause experiment......................................................................91 3-5 Mean and standard deviation for number of words for (a) Transitive responses in the Transitive experiment, (b) Dative responses in the Dative experiment, and (c) TwoClause responses in the Two-Clause experiment...............................................................91

PAGE 11

11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECT OF SYNTACTIC STRUCTURE ON SPEECH PRODUCTION IN ADULTS WHO STUTTER By Maisa A. Haj-Tas December, 2007 Chair: Kenneth J. Logan Major: Communication Sciences and Disorders This study examined (a) the effect of syntactic structure on speech fluency and speech reaction time (SRT) in adults who stutter, (b) the effect of synt actic priming on fluency in those participants. Fourteen persons who stutter (PWS) and fourt een persons who do not stutter (PWNS) participated in the study. A sentence st ructure priming paradigm was used to elicit sentences of varying syntactic forms (i.e ., transitive, dative, and two-clause). The results of the fluency analyses indicated that (a) PWS produced significantly more repetitions and prolongations per response than PWNS, (b) all participants produced a comparable number of other t ypes of disfluencies (e.g., interjections and revisions), (c) all participants seemed to produce more fluent re sponses following transiti ve pictures and fewer fluent responses following two-clause pictures. The results of the priming analyses indicated th at (a) the presence of passive primes did not significantly increase the probability of using passi ves in the response, (b) the structure of the dative primes affected the probabi lity of using those structures in the response differently between the study groups. Specifically, the PW S produced significantly more prepositionaldative responses following prepositional dati ve primes than following object-complement primes; however, no significant differences among the effect were observed in the PWNS. This

PAGE 12

12 finding was taken to suggest that priming dative sentence forms for the PWNS might be more driven by the structural aspects of the available options and for the PWS by the thematic aspects, and (c) the presence of the embedded primes di d not significantly increa se the probability of using embedded structures in the responses. The results of the SRT analyses indicated (a) no significant differences between the two groups in SRT, (b) that the syntactic structure of the response did not affect SRT differently between the two groups, and (c) a ll participants produced transitiv e responses at a significantly shorter SRT than dative responses, and dative re sponses at a significantl y shorter SRT than twoclause responses. The results of the SRT were taken to suggest that increasing the syntactic complexity of the responses may influence the time both the PWS and the PWNS may need to generate such responses.

PAGE 13

13 CHAPTER 1 INTRODUCTION The relationship between speech formulation pr ocesses and stuttering has been central to numerous hypotheses and investigations for severa l decades. A number of models of stuttering suggest stuttering is a problem that results from difficulties arising during speech formulation, that is, prior to speech production (e.g., Kolk & Postma, 1997; Perkins, Kent, & Curlee, 1991; Starkweather, 1987). Results of num erous studies also pr ovide evidence that increased linguistic, temporal, and cognitive loads seem to affect the speed and fluency of speech production in persons who stutter (PWS) more so than it doe s in persons who do not stutter (PWNS) (e.g., Bernstein Ratner & Sih, 1986; Bosshardt, 1993; Bosshardt & Fransen, 1996; Cuadrado & Weber-Fox, 2003; Logan & Conture, 1995, 1997; Logan, 2001, 2003; Silverman & Ratner, 1997; Yaruss, 1999). Despite all this evidence, th ere are still numerous unanswered questions about the role that speech formula tion plays in the disfluencies th at characterize stuttered speech. The present study was conducted to examine the role that syntax might play in speech production in general, and speech fluency and sp eech timing in specific in PWS. The plan of this chapter is as follows: Sect ion I provides an overview of a number of psycholinguistic models that offer different pers pectives about the cause of stuttering. Section II provides a review of empirical ev idence that supports the effect of certain linguistic variables, namely syntactic encoding, on the efficiency and timing of speech formulation processes in PWS. Section III presents a number of methodol ogical limitations associated with previous research that has explored the effect of synt ax on speech production processes in PWS. Section IV reviews a methodology (i.e., syntactic primi ng) that has been used to examine speech formulation processes in both speakers with typical language and sp eakers with atypical language functioning, and which has only been used infrequently to examine speech formulation

PAGE 14

14 processes in PWS. Finally, section V presents the rationale for using the syntactic priming methodology to examine the syntactic encoding abilities in and be tween PWS and PWNS. Psycholinguistic Models of Stuttering: An overview Numerous models of speech production sugge st that different aspects of speech formulation might be involved in the production of disfluenci es by PWS (for a review see Bernstein Ratner, 1997; Conture 2001 ). Such models seem to have a common underlying claim that disfluencies produced by PWS result from di sruptions in linguistic planning prior to speech production (Karniol, 1995; Kolk & Postma, 1997 ; Perkins, Kent, & Curlee, 1991; Smith & Kelly, 1997, Starkweather, 1987). In this section, I provide an overview of the five bestdeveloped models of speech production that resear chers have used in their investigations of linguistic planning abilities of PWS. The Covert Repair Hypothesis (CRH) Developed by Postma and Kolk (1993), this theory proposes that fluency breakdowns in stuttering result from difficulties that arise pr ior to speaking in genera l and at the phonological encoding level in specific. The CRH suggests th at although the self-rep air process and speech production monitoring skills in PWS are similar to those in typical speakers, the processes underlying the selection of phonemes or insertion of the selected phonemes into speech plans are slower or delayed in PWS compared to PWNS This slowness in phoneme activation or encoding results in particular phonemes being in competition for selection with other phonemes for a time frame longer than normal, which in turn increases the chance of phoneme selection errors if the speaker attempts to commence sp eaking before the phoneme selection process is complete. According to the CRH, stuttering arises when PW S detect such selection errors and try to correct them while speaking at a rate that is faster than their impaired phonological encoding

PAGE 15

15 mechanisms allow them to proceed. Within this view, the repetitions and prolongations observed in the speech of PWS are a result of attempts by those speakers to self repair the errors before they are spoken (i.e., covertly repair those errors). In Contur es (2001) review of the CRH, he explained that although this hypothesis has seve ral components that make it appealing (e.g., attempting to account for speech disfluencies that arise during conversational speech), the CRH does not account for other levels of speech formulation such as lexical/semantic and syntactic encoding. Thus, as Conture suggested, th e CRH may have limited ability to handle the sort of individual variations that other m odels so aptly attempt to describe. (p.37) The Neuropsycholinguistic Theory of Stuttering Developed by Perkins, Kent, and Curlee (1991), this theory suggests th at stuttering results from dyssynchrony in the integration of linguist ic units into their planned speech frames. Specifically, the theory suggest s that during speech formula tion, stuttering occurs if the generation of sounds or fillers and the integration of those fillers into syllable frames (or slots) become dyssynchronous. Perkins et al. (1991) ex plained that hesitatio ns during speech are expected to arise as a result of interactions between uncertaintie s about the planned linguistic unit and the need for the speaker to begin, con tinue, or accelerate an ut terance (p.734). Within such a framework, nonstuttered disfluencies such as revisions and interjections could occur when (a) a speaker is under time pressure and the cause of disfluency su ch as the speakers uncertainty about the linguistic characteristic s of the target is known, or (b) when the speaker does not know the cause of uncertainty but is not under ti me pressure. On the other hand, stutter-like disfluencies (e.g., repetitions and audible and in audible prolongations of sounds and syllables) may occur under the combination of two primary conditions (a) when disfluencies could arise from a variety of causes and the speaker is not ce rtain about which cause is contributing to them, and (b) if the speaker is under real or percei ved pressure to continue speaking when such

PAGE 16

16 disruptions occur. Within this framework, Perkins et al. (1991) define stu ttering as disruption of speech that is experienced by the speaker as a lo ss of control (p.734). Thus, time pressure is a central component in stuttering wi thin this theorys framework. The Suprasegmental Alignment Model of Stuttering Developed by Karniol (1995), this model suggests that stuttering results from problems at the speech formulation level. Unlike Kolk et al (1993) and Perkins et al.s (1991) models of stuttering, Karniol suggested that stuttering arises from difficulti es at the sentence level. The suprasegmental alignment model proposes that disfluencies produced by both PWS and PWNS result from the attempt of the speakers to revise sentence plans during online sentence production. Within such a perspective, Karniol e xplained that suprasegme ntal features of a sentence such as rhythm and stress are largely determined prior to utterance initiation and are expressed through changes in muscle moveme nts (p.111). Additionally, and as Karniol explained, although suprasegmental features are prep ared before the sentence is uttered, speakers often change their speech plans online. Such changes most often occur because speakers may initiate speech before the utterance planning pr ocess (especially that fo r the verb phrase) has been completed. If problems arise during speakin g and the utterance cannot be completed as originally planned, speakers may change those pl ans. Karniol suggests that both the latencies and hesitations observed in speech production reflect such sentence plan changes. In some sense, Karniols model is similar to Perkins et al.s mo del in that both propose that increased demands (such as the need to continue talking while attempting to correct selection errors under time pressure) are important contributors to the transformation of disfluenci es into stuttering events. The Demands and Capacities Model of Stuttering (DCM) This model provides yet another view of the pr ocesses that lead to stuttered speech (e.g. Starkweather, 1987). Similar to Perkins et al. neuropsycholinguistic model, the DCM suggests

PAGE 17

17 that increased demands affect speech production in PWS. The central claim in the DCM model is that stuttering (specifically in children) may re sult from demands within or between domains that exceed the capacities of the speaker to produce speech fluently (for a review and critique of the model, see Bernstein Ratner, 1997; Conture, 2001; Manning, 2000; Siegel, 2000, Yaruss, 1999, 2000). Much of the available critique of this mode l argues that although it is very appealing, the nature and level of demands are not well defined in the model. Some authors, however, provide their own interpretation of the claims of the DMC and suggest examples that may help better understand the nature of such demands. For example, in his review of this model, Conture (2001) provided one example to show how the assumptions of this model work during speech. In his example, Conture explained that although a sp eaker may have typical (or above average) capacities for generating utterances that are long or grammatica lly complex, these capacities might be challenged under some conditions, such as when the speaker attempts to rapidly produce long or grammatically complex uttera nces that may contai n complex articulatory adjustments. Within the framework of the DCM, such time pressure demands may exceed the speakers ability to make the speech motor adjustments needed to accurately and efficiently produce such utterances, and this in turn increases the probab ility of stuttering. The Multifactorial Model of Stuttering Developed by Smith and Kelly (1997), this m odel provides yet another view of the factors underlying stuttering and s uggests that stuttering results from factors that interact in a complex, nonlinear, and dynamic manner. Smith and Kelly expl ained that stuttered events should not be viewed as static behaviors that occur in isolati on at discrete points in time. Rather, stuttered events should be examined within a multifact orial framework in which interactions among various important factors are analyzed. The model al so suggests that defin itions and descriptions of stuttering need to take into ac count the reality that the nature of the interactions among factors

PAGE 18

18 will likely vary from one context to another a nd will also vary between speakers. The Smith et al. model seems to be similar to the other models re viewed in this section in the sense that all of the previously reviewed models s uggest that, in one way or anothe r, stuttered speech could result from interactions between two or more factors. The appeal of Smith and Kellys model is in the fact that it does not confine the cause of stuttering to just one factor and that it considers the variability of stutteri ng under different conditions. Also, a nd as Conture (2001) suggests in his review of this model, The Smith and Kelly mode l does a nice job of poin ting out that individual variations among different variab les may mean that we should be talking about causes rather than the cause of stuttering. (p.33). Although the models reviewed above adopt differe nt approaches in their attempt to explain the factors underlying stuttering, in one way or another, they all seem to suggest that stuttering might result from difficulties that arise dur ing language formulating and prior to speech production. As Conture (2001) suggeste d, such models may lead us to conclude that some of the factors underlying stuttering might be found between thought and motor execution of that thought (p. 354). Although other f actors such as motor execution have been implicated in stuttering, Conture suggested that the compone nts of speech production processes above the motor execution level such as semantic, phonol ogical, and syntactic components are fast enough and creative enough to account for what we know about stuttering (p.354). Indeed, numerous studies have examined the relationship between speech formulation processes and stuttering and results of several of those studies su ggest that such a relati onship might exist. The following section provides a review of those studies and the main findings they reported. In addition, the section provides evidence from stud ies with both children and adults who stutter

PAGE 19

19 about the effect of different li nguistic factors in general and the effect of grammatical or syntactic complexity in specific on speech fluency in those speakers. Linguistic Factors and Stuttering Empirical evidence from numerous studies has consistently shown that mechanisms underlying language formulation processes such as phonological, semantic, and syntactic encoding may operate atypically in PWS compared to their non stuttering peers (Bernstein Ratner, 1997; Bosshardt, 1993; Bosshardt & Fransen, 1996; Cuadrado & Weber-Fox, 2003; Kolk & Postma, 1997; Wijnen & Boers, 1994; Weber-Fox, Spencer, Spruil, and Smith, 2004). Although on one hand results of several such st udies suggest that PWS on average perform within the normal range on static, offline m easures of linguistic abilities (e.g., Nippold,1990; Watkins, Yairi, & Ambsrose, 1999), on the other ha nd, results of numerous studies suggest that the combination of a fragile linguistic system and a variety of linguistic and cognitive factors may result in subtle differences between PWS and PWNS during online speech processing (Bernstein Ratner, 1997; Bossh ardt, 1993; Bosshardt & Fransen, 1996; Cuadrado & Weber-Fox, 2003; Weber-Fox, Spencer, Spruil, & Smith, 2004). Although different linguistic processes involve d in speech production in PWS have been examined thus far, the contribution of grammatical or syntactic complexity has received most of the attention during the past years. Specificall y, many investigations have been conducted to examine the nature of syntactic events that cont ribute to stuttering and th e effect of manipulating such events on speech fluency and accuracy in PW S. In the following section, a brief review will be provided of the main findi ngs of such investigations. Research on the location and frequency of disfluencies within utterances produced by speakers who stutter indicate th at disfluent events do not occur randomly during speech. Specifically, and as Peters and Guitar (1991) expl ained, Stuttering occurs more often at points

PAGE 20

20 in the utterance that can be described in lingu istic terms. (p.117). For example, results with PWS have consistently shown th at stuttering tends to be obser ved more often on (a) sounds in the word-initial position rather th an sounds in other word positi ons, (b) consonants rather than vowels, (c) multisyllable words rather than m onosyllable words, (d) content words (e.g., nouns, verbs, adjectives, adverbs) ra ther than function words (e.g., ar ticles, prepositions, and pronouns), (e) stressed syllables rather than unstressed syll ables, (f) low frequency words rather than high frequency words, and (f) words occurring at or near the sentence initial position rather than sentence final position (for more information, see reviews in Bloodstei n, 1995; Bernstein Ratner, 1997; Peters & Guitar, 1991). Results of numerous studies have also shown th at in children who stutter (CWS), stuttering is more likely to occur within utterances that are long and/ or syntactically complex than utterances that are s hort and/or syntactically simple (e .g., Bernstein Ratner and Sih, 1987; Gaines, Runyan, & Meyers, 1991; Gordon, Luper, & Peterson, 1981; Logan & Conture, 1995; 1997; Wall, Starkweather, & Cairns, 1981; Yaruss, 1999). In general, these results have been interpreted within theoretical frameworks such as the ones reviewed earlie r in this chapter. For example, within the framework of the DCM, incr eased utterance length a nd syntactic complexity may be viewed as demands that exceed the child s capacity to produce such utterances fluently (e.g., Logan & Conture, 1997). Although, as mentioned above, lingu istic complexity has been shown to affect fluency in CWS, investigations of this effect in older sp eakers who stutter provide different results (e.g., Logan, 2001; Silverman & Bernstein Ratner, 1997) For example, Silverman and Ratner (1997) examined the relationship between syntactic comp lexity and fluency in adolescents who stutter and adolescents who do not stutter. The participants heard and th en repeated utterances which

PAGE 21

21 featured one of three levels of grammatical complexity (i.e., Wh questions, right-embedded relative sentences, and center embedded relative se ntences). The authors measured the effect of utterance complexity on the accuracy and fluenc y of sentence imitation by the two groups. The results of the study indicated that increased grammatical complexity affected both speaker groups in a similar manner. Specifically, as the gr ammatical complexity of the target utterance increased, the percentage of disf luencies and inaccurate response s also increased in both groups. Based on the results of their study, Silverman et al. (1997) conclude d that although some syntactic structures continue to be challenging for both adolesce nts who stutter and adolescents who do not stutter (as observed in the comparable increase in speech disf luencies in both groups in the more syntactically challe nging sentences), the effect of syntactic complexity appears to diminish as speakers grow older and th eir linguistic competency improves. In a similar investigation, Logan (2001) st udied the effect of increased syntactic complexity on speech fluency and timing in younge r (i.e., adolescents who stutter) and older PWS as compared across two speaking contexts: (a) a conversational spee ch task during which the participants generated utterances during a st ructured conversation, an d (b) a sentence reading task that was performed within a reaction time paradigm. In the sentence reading task, the participants read and rehearsed a series of sentences in which the subject noun phrase differed in syntactic complexity. In detail, Logan prepared a base level sentence (e.g., The cake was served at the dinner) and then generated four sentence versions from the base sentence by changing the structure in the subject constituent of the se ntence. As Logan explained, in sentence version one, a Determiner + Adjective + Adjective + Noun subject constituent was used (e.g., The rich and tasty cake was served at the dinner ). In sentence version two, the subject constituent was Determiner + Noun + Prepositional Phrase stru cture (e.g., The cake from Bills oven was served

PAGE 22

22 at the dinner). In sentence version three, the subject constituent was Determiner + Noun + Relative Clause (The cake that was fancy was serv ed at the dinner.). Fina lly, in sentence version four, the subject constituent was Determiner + Noun (e.g., The cake was served at the dinner for the mayor) (For more details, see Logan, 2001, pp. 92-93). Participants were presented with and asked to reproduce the sentences initiating th em as promptly as possible following the presentation of a cue (i.e., 1 KHz a pure tone ). Consistent with the findings reported by Silverman et al. (1997), Logan reported that sp eech fluency is comparable between the two groups when producing sentences of varying synt actic complexity. In other words, different levels of syntactic complexity seemed to sim ilarly affect speech fluency in adolescents and adults who stutter. Logan also reported that the means for the different disfluency types examined in the study were similar for the tw o groups. Based on these finding, Logan concluded that younger PWS seem to control their level of speech fluency and manage sentences of varying grammatical complexity in a manner different from that used by older PWS. Based on those results, Logan suggested, in agreement with Silv erman et al., that th e factors that underlie stuttering in children might not persist beyond the childhood years. Methodological Limitations Although the results from Silverman et al. ( 1997), and Logan (2001) may appear to argue against a role for syntactic complexity in stutte ring in adolescents and adults who stutter, the authors of both studies suggested that using speech elicitation pa radigms other than the ones used in their studies (i.e., paradigms other than sent ence repetition) might pr ovide information about subtle syntactic effects that exist in the speech production process of PWS. For instance, Logan noted that, in his study, syntactic complexity was defined within a transformational grammar framework. He also suggested that alternate fram eworks of complexity, such as those based upon memory cost models, might yield different results.

PAGE 23

23 Indeed, a number of alternative paradigms examining the relationship between syntactic complexity and stuttering have provided a differe nt view of this relati onship. Such alternative paradigms include using dual-ta sk activities (e.g., Bosshardt, Ballmer, & De Nil, 2002), online grammaticality judgments (e.g., Cuadrado & Webe r-Fox, 2003), and speech reaction time (SRT) (e.g., Logan, 2003). Although the three above paradigms are di fferent in terms of their procedural details, the motivation underlying thei r use seems to be similar in that they all measure the effect of syntactic complexity on speech production in indi viduals who stutter under conditions of increased cognitive and time load s during online investigations of the speech production process. As suggested by Cuadra do and Webber-Fox (2003), Additional cognitive loads (e.g., in a dual-task paradigm) have been shown to accentuate differences between NS [non stutterers] and IWS [individuals who stutter] in planning of linguistic units in speech production. (p. 962). The results of such studies have consistently shown that although there is an overlap in the response fluency between PWS and PWNS, increased cognitive loads and interfering attention-demanding tasks seem to have a more pronounced effect on the efficiency of retrieval and or encoding of linguistic units during sentence fo rmulation in PWS. For example, results of several studies revealed that under dua l tasks (compared to single tasks), PWS require more time for sentence generation and articulati on and produce more speech disfluencies than PWNS (Bosshardt, Ballmer, & De Nil, 2002; Caruso, ChodzkoZajko, Bidinger, & Sommers, 1994). Similarly, results of other studies rev ealed that under more demanding online tasks (i.e., tasks that are used to measure the effects of a given variable on speech production during ongoing speech processing) versus offline tasks (i .e., tasks that are used to measure the effects of a given variable on speech production after or at the end-point of speech processing), the accuracy of grammaticality judgment in general a nd syntactic processing in specific in PWS are

PAGE 24

24 most affected by the increased length and syntac tic complexity of the target sentence during online tasks as compared to offline tasks (e.g., Cuadrado & Weber-Fox, 2003). As mentioned above, although syntactic comp lexity has repeatedly been shown to contribute to breakdowns in fluency in CWS, in vestigations of this linguistic phenomenon in older PWS seem to provide mixed results. For example, Silverman et al. (1997) and Logan (2001) suggest that the effect of syntactic complexity on fluency might be minimal in older speakers who stutter, while Bossh ard (1993) and Cuadrado et al. ( 2003) suggest that this effect can be observed under certain cognitive demands. The difference between the results offered by the these two study groups could be attribut ed to several factors among which are the methodological differences. The later group of studi es has examined the effect of syntactic complexity on speech formulation and production abilities of PWS wh en a concurrent or secondary task was used. The results of those studies show a significan t effect of linguistic phenomena in PWS. Thus, methodologi es that used concurrent tasks seem to add to the cognitive loads during those experiments and that in turn offered a glimpse into the dynamics of this interaction The following section provides an overview of an alternative online, and thus more demanding methodologysyntactic priming. This methodology has been used for several decades in psycholinguistics research to examin e speech formulation processes in both typical speakers and speakers with language delays. As will be shown below, and in agreement with several other researchers, I believe that s yntactic priming is a potentially informative methodology for examining syntacti c encoding abilities of PWS. The Syntactic Priming Methodology Many observations of natural language have re ported that speech is highly repetitious (e.g., Levelt & Kelter 1982; Bock, 1986). For mo re than two decades, researchers have successfully shown that the repe titious nature of language can be induced under experimental

PAGE 25

25 conditions by using the priming methodology. The fo llowing section will provide an overview of the priming methodology in general, and the sy ntactic priming methodology in specific. In addition, the following section will provide an overvi ew of the studies that have used the priming methodology to examine the speech formulatio n processes in speakers who stutter, and particularly, the syntactic pro cessing abilities in this populat ion. A review of phonological and lexical/semantic priming studies will be included for the sake of establis hing the feasibility of this methodology in exploring the speech form ulation and production processes in PWS. Syntactic Priming and Speakers with Typical Language Production It is well documented that, during conversat ions, speakers tend to repeat grammatical structures that either they or their interlocutors produced earlier in those conversations (for a review, see Bock, 1986; Pickering & Branigan, 1 999; Branigan, Pickerin g, McLean, & Stewart, 2006; Smith & Wheeldon, 2001). As Smith and Wheel don (2001) explained, repetition in natural language seems to affect all aspects of the gramma tical structure of langua ge such as syntactic frames, formulaic language, individual lexemes or phases, or the lexico n itself(p.124). Bock (1986) was among the first researchers to show that the repetition of one particular grammatical structure that has been observed in natural language can also be observed in experimental settings. This phenomenon has been termed, va riably, structural pe rsistence, syntactic persistence, sentence structur e priming or syntactic priming The term syntactic priming will be used in the present study. Several hypotheses have been proposed to e xplain the syntactic priming phenomenon. In her pioneering examination of this phenomenon, Bock explained that her studies were designed to explore The syntactic activation hypothesis, which examines grammatical patterns in speech result from the application of cognitive realizations of syntac tic rules or structure heuristics, perhaps via procedural representati ons of grammatical structures (p. 358). Bock

PAGE 26

26 suggested that one consequence of such an application could be a strengthened or elevated level of activation of certain rules or structures, whic h result in the repetiti on of a given syntactic structure. To explore this hypothesi s, Bock prepared sentences that contained syntactic structures reported in earlier observations to be associated with the persiste nce (repetitious) effect such as dative sentence structures (e.g., The boy gave th e girl a valentine) and passive sentence structures (e.g., The man was stung by a bee). During the experiment, Bock presented participan ts with sentences and pictures that were transitive (i.e., the main in those sentences and pictures requires an obj ect as in The fireman rescued the baby) and dative (i.e., the main verb in those sentences and pictures requires two objects as in The boy(1) gave the girl(2) a valentine). The participants first heard then repeated sentences such as the transitive sentence One of the fans punched the referee. Next they saw a picture such as that of a tornado destroying a ba rn and were asked to briefly describe what was happening in that picture in one complete se ntence without using any pronouns. The design of the study was such that each of the transitive ta rget pictures could be described by either an active sentence (e.g., A tornado destroyed the ba rn) or a corresponding passive (e.g., The barn was destroyed by a tornado). Similarly, each of the dative pictures (e.g., a boy giving a valentine card to a girl) could be described either by a pr epositional dative sentence (e.g., The boy gave a valentine to the girl) or a double object dative sentence (e.g., The boy gave the girl a valentine). The results of the study showed that the presentation of a certa in syntactic structure (i.e., syntactic prime) prior to a target picture increased the probability of that structure being used to describe the target picture. For example, participants seemed to show an increased use of active structures to describe a picture when the prim ing sentence they repeated was of an active structure. In a series of follo w-up experiments, Bock and colleagues provided evidence that the

PAGE 27

27 generated syntactic structures were not a result of conceptual characteristics of the sentences, and that the priming effect was obser ved regardless of the content of those priming sentences (Bock, 1986; Bock and Loebell, 1990). Based on the re sults of those studies, Bock and colleagues concluded that the structural pe rsistence (i.e., priming) effect observed in their experiments was actually a result of activation of syntactic processes, wh ich in their opinion, supported the above mentioned syntactic activation hypothesis. Bock (1986) also suggested th at although processes such as priming may limit the fl exibility of syntax in spontaneous real time use of language, perhaps contributing to differences in syntactic diversity between planned and extemporaneous speech, (p.379), such a process could also have an adaptive function. Bock further explained that the expression of unplanned messages during speech can create problems that lead to hesitations, errors, and other di sruptions (p.379). As suggested by Bock, reusing structures that were available in an earlier sent ence (within a context similar to th e one used in the study) could contribute to fluency since it em ploys procedures that are alrea dy activated and this in turn may ease the demands of generating the target message. These conclusions were tested more recently in a series of six experiments conducted by Smith and Wheeldon (2001). The experiments examined the effect of syntactic priming on speech initiation times (SIT), am ong other variables, in typical adult speakers. The authors suggested that if syntactic persistence result s from reusing already activated procedures, syntactic priming may positively affect tempor al factors in speech production by reducing the time speakers need to initiate utterances. Th e first experiment reported by Smith et al. (2001) was designed to investigate the hypothesis that syntact ic persistence reduces the time dedicated to syntactic planning (as observed in reduced sp eech onset latencies or a shorter SIT). The material of the study consisted of a variety of moving pictures. In the s yntactically related

PAGE 28

28 condition and as described by Smith et al. the prime and target trial sentences were matched both in terms of the grammatical role of the phrases they assigned nouns to (i.e., a subject phrase) and in terms of the comple xity of the internal structure of the phrase they assigned nouns to (i.e., a conjoined noun phrase.(p.130). An exam ple of a target phrase used in the study is The spoon and the car move up, and a corresponding syntactically related sentence is The eye and the fish move apart. In the syntacti cally unrelated condition, Smith and Wheeldon (2001) explained that the prime trials matched the target trials in terms of the grammatical role of the phrases they assigned nouns to but not in terms of the complexity of the internal structure of the phrase they assigned nouns to (i.e., a simple noun phrase rather than a conjoined noun phrase). (p.130). An example of a syntactically unrelated priming sentence used in the study is The eye moves up and the fish moves down. During the experiment, participants describe d black and white line drawings of objects moving in different directions on a computer screen. Prior to th e experiment, the participants were instructed to describe the picture moveme nts from left to right using specific sentence types. Speech production latencies were measured to determine whether there were significant differences in speech latencies between the synt actically related and sy ntactically unrelated conditions. The results of the study revealed that target sent ences that were preceded by syntactically related primes were produced at significantly shorte r (55 ms) speech latency than target sentences that were preceded by unrelated syntactic primes. Based on the results of the study, Smith and Wheeldon concluded that Clearl y, such a result provid es the first evidence compatible with the hypothesis that syntactic persistence benefits speakers by reducing the processing costs incurred by syntac tic structure generation (p.138).

PAGE 29

29 As observed in both Bock (1986) and Smith and Wheeldon (2001) experiments, syntactic priming seems to be a reliable methodology for examining sentence structure generation and production processes in typical speakers. In addi tion, the presentation of syntactically related primes prior to targets seems to facilitate sentence generation processes as measured by the reduced speech latencies under the related synt actic prime condition. In an above-presented explanation of how the syntactic effect works, Bock had suggested that although syntactic persistence might limit the flexib ility of syntax in spontaneous speech it might also have an adaptive function in the sense that the expres sion of unplanned messages during speech can create problems that lead to hesitations, erro rs, and other disruptions (p.379). Both Bock (1986) and Smith et al. (2001) also seem to agree th at message generation under some conditions might impose additional cognitive loads that may affect the timing and fluency of the uttered sentence. Since syntactic priming has been shown to enhance fluency during real time tasks in typical speakers, numerous studies have examined this phenomenon and whether similar results can be observed with atypical speakers. The fo llowing section provides re sults of such studies. Syntactic Priming and Speakers with Atypical Language Production In addition to its use with speakers exhibi ting typical language ab ilities, the priming methodology has also been used to examine synt actic processing abili ties in speakers with atypical language production. For example, the syntactic priming paradigm has been used to examine difficulties in grammatical processing in aphasic patients (e.g., Hartsuiker & Kolk, 1998; de Roo, Kolk, & Hofstede, 2003), patients diagnosed with dyslexia (Faust, Silber, & Kaniel, 2001), and patients diagnosed with schizophrenia (e.g., Besche, Passerieux, Segui, Sarfati, Laurent, & Hardy-Bayle, 1997). For example, Hartsuiker et al.s 1998) study re vealed that patients with Brocas aphasia showed stronger syntactic priming effects than th e typical control speaker s, and that although the

PAGE 30

30 speech of such patients is usua lly characterized by a reduced comp lexity of syntactic structure, these patients produced relatively complex senten ce structures (such as passives) when those structures were primed. Hartsuik er et al. (1998) expl ained that some hypotheses suggest that patients with Brocas aphasia might suffer from reduction in their capacity for generating grammatical structures, especia lly complex ones, which may result from either slowness or reduced maintenance/fast decay in the genera tion of grammatical information. Within this view, and based on the results of their study, Hartsu iker et al. provided a number of conclusions among which was that limitations in computationa l resources in patients with Brocas aphasia can be overcome by using facilitatory processes such as syntactic priming. Besche et al. (1997) used syntactic priming to examine the hypothesis that some types of schizophrenic disorders might have the disorga nization of cognitive acti vities as an underlying cause and particularly disorganization of seman tic and syntactic relations during the processing of lexical information. The results of the study revealed that the performance of the schizophrenic patients pertaining to the processing of syntactic in formation was identical to that of the typical controls In addition, results revealed that both groups showed a significant syntactic priming effect as observed in the fo rm of reduced SIT during a decision making task. As can be concluded from the two examples above, syntactic priming is a methodology that can be helpful in examining syntactic fo rmulation and production ab ilities in atypical populations as well as typical sp eakers. The following section provides an overview of a number of studies that have used the priming methodol ogy to examine speech processing abilities of PWS. The Priming Methodology and People who Stutter As mentioned above, both empirical evidence an d theoretical explanations suggest that stuttering might result from difficulties in genera ting or encoding syntacti c units. In addition, a

PAGE 31

31 number of hypotheses suggest that speakers who stutter may exhibit limited capacities to generate fluent speech under c onditions of increased linguistic or cognitive demands (DCM; e.g., Starkweather, 1987). If this is the case, then the facilita tory effect of syntactic priming that has been shown in several clinical populations who exhibit limitations in speech generation capacities (e.g., patients with Bro cas aphasia) may also enhance speech fluency in PWS. Although the priming methodology has been used in studies of speech production with both typical and clinical populations as shown in the previous section, its application to stuttering is still in its infancy. Most of the studies that have used priming to examine speech formulation in PWS have used priming to examine the ti me course of phonological and lexical/semantic encoding in children and PWS (e.g., Burger & Wijnen, 1999; Hartfiel d & Conture, 2006; Melnick, Conture, & Ohde, 2003; Pellowski & C onture, 2005; Wijnen & Boers, 1994). To my knowledge, only one study thus far used priming to examine the syntactic processing abilities of speakers who stutter (Anderson & Conture, 2004). Although as mentioned syntactic priming is the scope of the present study only covers syntactic priming, I believe that is it worth presenting studies that used other primi ng methodologies such as phonological and lexical/semantic priming in persons who stutter. Presenting such studies will help establish that the priming methodology can be reliably used to examine speech formula tion and production abilities of this population. The following section first provides an overvie w of studies that have used the priming methodology with PWS and the main results of thos e studies, and then provides the rationale for using syntactic priming to examine synt actic processing abi lities of PWS. Phonological Priming and PWS Wijnen and Boers (1994) were the first re searchers to use the priming methodology to examine phonologic encoding abilities of PWS. Th e study was designed to test the hypothesis that stuttering results from problems in phonological encoding a nd particularly the encoding of

PAGE 32

32 syllable rime (the syllable constituent capturi ng the vowel and any final consonants (p.5)). Wijnen and Boers explained that It is assumed th at priming increases the activation level of the involved phonological segment(s) so that incorporation [of those segments] in the articulatory plan is facilitated. (p.6). The participants in the study were ni ne adults who stutter and nine adults who do not stutter (all native speak ers of Dutch). Two phonological priming conditions were used: priming the initial consonant (C-B lock), and priming the initial consonant and subsequent vowel (CV-Block). A cue word was visually presented and the participants had to utter a response word (out of a set of five wo rds) that was assigned to the cue word and which the participants learned prior to the online experi ment. There were five blocks of five-word sets that were either homogeneous (all words shar ed the same consonant or shared the same consonant and vowel) or heterogeneous (wor ds were a combination from each of the homogeneous sets). For example, in the homoge neous condition of a C-Block, the cue word was vork the corresponding response words were Lepel, Lila, l oeder, larie, and Luir. Speech reaction time was measured in each of th e priming conditions. The results of the study indicated that SRT in the C-Block was signi ficantly faster during the homogeneous condition (606 ms) than it was in the heterogeneous conditi on (639 ms), indicating a larger priming effect during the former condition (although, as the au thors reported in a follow up study presented below, this effect was only obser ved in only four out of the nine study participants). The authors also reported that in most pa rticipants who stutter, reducti on of SRT occurred only when the words shared both the consonant and vowel. Base d on the results of the study, Wijnen and Boers concluded that there is a diffe rence in phonological encoding ab ilities between PWS and PWNS, and that the results indicate th at the encoding of stressed vowel at the non-initia l parts of the syllable might be delayed PWS.

PAGE 33

33 In a follow up study, Burger and Wijnen (1999) replicated the Boers et al. study using a larger group of participants and new stimulus wo rds. Similar to Boers et al., the results of the follow up study indicated that (a) the overall SRT for PWS was slower than that for PWNS, (b) SRTs were faster in the homogeneous condition compared to the heterogeneous condition, and (c) the priming effect was more prominent when the priming word and the target shared the initial CV than when they only shared the initial C. However, Burger et al. also reported that they failed to replicate the interact ion effect between group, prime type, and condition reported in the Boers et al. study. In other words, the authors re ported that both groups benefited from priming similarly and showed a similar and larger effect of priming in the CV condition compared to the C condition. In a more recent study, Melnick and Conture (20 03) also examined the effect of priming on phonological encoding in general and speech reacti on time in particular in preschool children who do and do not stutter. The authors hypothesized that if CWS do, in fact have an impaired ability to quickly phonologically encode as the CRH suggests, then their ability to benefit from phonological priming may not be as a pparent or as great as that of CWNS. (p. 1438). The study included 18 children who stutter (CWS) and 18 children who do not stutter (CWNS) (3-5 year old). Unlike Wijnen and Boerss (1994) study design, Melnick and Cont ures (2003) priming condition had three levels: No prime, related phonological prime, and non related phonological prime. In the no prime condition, the children saw pictures of common objects and were asked to name the objects as fast as they could. In th e related priming conditi on, the children saw and described the same objects presented in the firs t condition; however, 500 ms prior to picture presentation, the children heard either a relate d or unrelated auditory stimulus. The auditory stimulus either shared (related prime) or did no t share (unrelated prime) the first CV or CCV of

PAGE 34

34 the picture (authors did not pr ovide examples). Participants were instructed to name pictures as fast as they could and SRT was recorded during the trials. Similar to the results reported by Wijnen and Boers (1994), the results revealed th at children in both groups exhibited faster SRT during the related versus unrelated prime conditi ons. The results also indicated that the older children in both groups exhibited an overall fa ster SRT than the younger children, and that the older children in both groups exhibited faster SR T during the related priming condition than the two other conditions. Post hoc analyses revealed that there was a difference between the two groups in the relationship between childrens scor es on tests of articulatory mastery and their reaction time scores. Specifically, the authors reported that the hi gher the scores of the children who do not stutter on articulatory mastery tests, the faster their react ion times were during picture naming. This relationship was not observed in CWS. Based on these results, Melnick and Conture concluded that presc hool children who stutter, as a group, may have somewhat less well developed articulatory systems than presch ool children who do not stutter. (p.1428). Most importantly, the authors suggested that phonol ogical priming methodology could help in examining the speech-language planning and produc tion of young (i.e., 3-5) year-old speakers. Lexical/Semantic Priming and PWS Priming methodology has also been recently us ed to examine lexical/semantic-encoding abilities of CWS (Hartfield & Conture, 2006; Pellowski & C onture, 2005). Pellowski and Conture examined the effect of priming on SRT in both CWS and CWNS. Th e participants in the study were 23 CWS and 23 CWNS st utter ranging in age from 3;0 (years;months) to 5; 11. The study included three priming conditions: no prime (i.e., no auditory stimulus was presented prior to the target picture), a relate d prime condition (i.e., a semantica lly related prime was presented 700 ms prior to the target picture; e.g., auditory prime boat and target picture car), and a non related prime condition (i.e., a se mantically unrelated prime was presented 700 ms prior to the

PAGE 35

35 target picture; e.g., auditory prime fork and target picture car). The authors reported two findings pertaining to the priming effect: (a) SRT was significantly slower in CWS compared to CWNS across the three primi ng conditions, and (b) When co mparing no prime condition to semantically related condition, resp onses of CWS were significantly slower than the responses of CWNS. Based on these results, Pellowski and Conture concluded that CWS may exhibit difficulties with lexical encoding which in turn may contribute to the speech disfluencies produced by those children. Similar to Pellowski et al. ( 2005), Hartfield and Conture (200 6) examined the effect of conceptual and perceptual pr operties of words on lexical retrieval in CWS and CWNS. The participants in the study were 15 3-5 year old CWS and a matche d group of CWNS. Participants in the study named objects presented under four le xical priming conditions: (a) a neutral primepure tone, (b) a prime word physically related to the target word (e.g., prime word tomato and target word apple), (c) a prime word functiona lly related to the target word (e.g., prime word bite and target word apple), and (d) a prime word categorically related to the prime word (e.g., prime word lemon and target word appl e). The authors reported the three following main findings: (a) Similar to the findings of Pe llowski et al., CWS were significantly slower at initiating fluent accurate responses than CWNS (b) Functionally related prime and target resulted in faster responses from CWS (but not CWNS) than physically re lated prime and target, and (c) there were no significant difference s between CWS and CWNS in error production. Based on the results of the study, Hartfield and Conture suggested that the difference between the two groups in SRT during picture naming c ould result from subtle differences in various aspects of speech-language planning (p.320). The authors further suggested that the source of

PAGE 36

36 these differences might be found in (1) the lingui stic plan, (2) the motor program, and/or (3) the transfer of information between lingui stic plan and motor program (p. 320). Syntactic Priming and PWS To my knowledge, Anderson and Conture (2004 ) are the only researchers who used the priming paradigm to examine syntactic processing abilities in persons who stutter. As Anderson et al. (2004) explained, their study was motivated by consistent earlier reports of relationships between utterance length, syntactic complexity, a nd fluency in CWS. The authors explained that findings of such relationships suggest that young CWS may experience some degree of difficulty quickly and/or efficiently formul ating morphosyntactic structures (p.553). The authors suggested that using the syntactic prim ing methodology may help us better understand selective aspects of the temporal component of linguistic pr ocessing in young CWS, particularly those associated with syntact ic processing (p. 555). In detail, Anderson and Conture used th e syntactic priming methodology to examine syntactic processing in CWS. Th e participants in the study we re 16 CWS and 16 CWNS stutter between the ages of 3;3 and 5;5. Th e authors used a syntac tic priming task that they described as an age-appropriate version of the syntactic priming paradigm [which was used by Bock and colleagues in several psycholingu istic studies] (p.552). The author s explained that latencies of responses (i.e., SRT) for the participating children were meas ured during a syntactic priming task that consisted of two conditi ons-a no prime condition and a syntactic prime condition. An unrelated syntactic prime conditi on was not used in the study because a pilot study showed that when such a condition was used with younger childre n, the unrelated primes seemed to affect the probability of the children later producing the target syntactic structures and thus resulting in the children not producing enough samples of the target sentence structure.

PAGE 37

37 The experimental pictures used in the study could be described by using simple, active, affirmative, declarative (SAAD) sentences, such as The girl is petting the cat. The other/filler pictures could be described usi ng other types of sentences such as negatives. In the no-prime condition, children were presented with pictures a nd were instructed to describe each picture as quickly as possible. Each picture was presen ted for 3000 ms, and childrens responses were recorded so that SRT could be assessed for each of the responses. During the no-prime condition, children saw the same pictures used in the s yntactic prime condition however, 2000 ms prior to the onset of picture presentation, a priming sentence that exhibited a syntactic structure similar to that of the target sentences was auditorily presented. Consistent with studies of typical speakers (e.g., Smith & Wheeldon, 2001), the results of the study revealed that SRT for both groups wa s shorter during the re lated syntactic prime condition compared to the no-prime condition (in CWS: shorter by approximately 212 ms, and in CWNS: shorter by approximately 51 ms). Results further revealed that there were significant differences between the two groups in the prim ing effect. Specifically, CWS were significantly faster by approximately 212 ms in the syntactic -priming condition than in the no-prime condition compared to CWNS (51 ms) who did not show a significant difference in SRT between the two conditions. Based on these re sults, and among other conclu sions, Anderson and Conture concluded that CWS seemed to benefit mo re from syntactic priming than CWNS. Summary In summary, both theoretical and empirical ev idence suggest that st uttering might result from difficulties in speech formulation processes prior to speech production. In addition, evidence from numerous studies suggests that increased linguistic loads during the speech production tasks seem to affect the temporal and fluency characteristics in PWS differently than in PWNS. Experimental designs that incorporate methodologies such as priming may offer a

PAGE 38

38 better understanding of how increased linguisti c demands affect speech formulation and production in PWS. Methodologies such as the s yntactic priming methodology and the material used by Bock and colleagues during the study task s could be specifically helpful in such an investigation because this material has the following appealing features: The material used by Bock and colleagues in the syntactic priming studies has been used for over two decades and has provided reliable results about syntactic processing in typical adult speakers. This in turn is expected to pr ovide us with a basis for comparing the results we obtain in the present study with both PWS and PWNS to the results of earlier studies that used this methodology. A series of follow-up studies by Bock and coll eagues clearly established that the carefully prepared material which was used in the syntact ic priming studies insu red that the syntactic priming effect observed in studies using that material is independent of lexical (Bock, 1986) and thematic/semantic e ffects (Bock & Leobell, 1990). A more recent study with CWS (Anderson a nd Conture, 2004) show ed that syntactic priming is a feasible procedure for studying speech-language planning and production in PWS. The sentences used by Bock and colleagues in the syntactic priming studies include sentences that cover a variety of syntactic st ructures (e.g., transitive, dative, two-clause embedded and conjoined) and syntactic co mplexities (i.e., transitive-low, dativeintermediate, and two-clausehi gh syntactic complex ity) and thus can be used to examine the effect of different sentence types on speech production processes in both PWS and PWNS (a detailed review of those sentence structures is presen ted in Chapter II). Thus, it seems reasonable to suggest that using the syntactic priming methodology, particularly the materials used by Bock and coll eagues in their syntactic priming studies, could provide important information and initial insi ghts into the temporal and syntactic production abilities of persons who stutter and in turn yield useful inform ation about the factors that are associated with disfluent speech. Research Questions The central question for this study is as follo ws: Is there a significant difference between adults who stutter and adults who do not stutte r in syntactic formula tion and production under conditions of increased linguistic demands? The four following questions are related to this

PAGE 39

39 broad issue and are specifically re lated to the effect of syntactic structure and syntactic priming on speech fluency and speech reaction time in responses produced by PWS and PWNS. Question One Does syntactic structure affect speech fluenc y of persons who stutter and persons who do not stutter? If so, is speech fluency of persons who stutter affected differently by syntactic structure than that of pe rsons who do not stutter? Question Two Does syntactic structure affect speech reac tion time in persons who stutter and persons who do not stutter? If so, are sp eech reaction times of persons w ho stutter affected differently by syntactic structure than those of persons who do not stutter? Question Three Does syntactic priming affect speech fluenc y in persons who stutter and persons who do not stutter? If so, is the percentage of fluent responses produced by persons who stutter affected differently by syntactic priming than that of persons who do not stutter? Question Four Does syntactic priming affect speech reaction time in persons who stutter and persons who do not stutter? If so, are speech reaction times of persons who stutter affected differently by syntactic priming than those of persons who do not stutter?

PAGE 40

40 CHAPTER 2 METHODS Participants 15 PWS and 15 PWNS participated in the study. The mean age for the participants who do not stutter was 32.2 years (S.D. = 12.76 years; range 21 59) and the mean age for the participants who stutter was 32.4 years (S.D. = 14.05 years; range 19 59). Persons in the two groups were matched for years in school (+/3 ye ars), age (+/2 years) and gender (10 male, and 4 female participants in each group). Overall, 14 out of the resu lting participant pairs were also matched for age and years in school. One pair was not matched for years in school nor age (PWS: 12 years in school, 52 years old; PWNS: 16 years in school, 24 years old). Several mechanisms were used to recruit participants These included contacti ng patients on the waiting list for diagnostic evaluations at the University of Florida Speech and Hearing Clinic and at several speech and hearing clinics in the State of Florida, posting signs on the University of Florida campus, and by making announcements in uni versity classes, and posting advertisements in local and college newspapers. Prior to participating in the study, participants completed several formal and informal tasks that were used to determine their eligibility fo r the study. These tasks are explained in detail below. Participant characteristics are presented and summarized in Table 1. Inclusion and Exclusion Criteria Participant inclusion and exclusion criteria were as follows: General inclusion criteria: To be incl uded in the study, participants had to: be native speakers of American English. have a negative history of any medical, neurological, or em otional conditions that might influence their performance during the study.

PAGE 41

41 have no active speech or language impairments other than stuttering and have no past history of language-related sp ecial services at any point during their academic careers. There were 16 participants who stutter and 22 participants who do not stutter who initially participated in the study. From this initial group, 3 PWS and 8 PWNS were excluded from the analysis for the following reasons: One participant who stutters completed the protocol for the study; however, the participants data were excluded from final anal ysis because he failed to meet the inclusion criteria described above (i.e., had a pos itive history of neurological problems). Although 22 participants who do not stutter comp leted the study protoc ol, data for only 15 participants in this group was used in the analyses. These participants were selected to insure a good age, education, and gender match with the participants who do not stutter. The excluded participants either did not match the PWS in age or in education. In addition, several of the participants in the PWNS group had an academic background in linguistics and data from meetings with those participan ts were not included fr om the analyses. This was done to eliminate any bias in the result s based on those participants prior knowledge or exposure to procedures and data simila r to the ones used in the present study. Material Measurements of speech reaction times, speech fluency, and sentence structure were made from audio and video recordings, which were co llected during a syntac tic priming task that lasted approximately 50 minutes. Details about th e study material (i.e., pre-study tasks and main task), data preparation, and study procedures are described immediately below and are also summarized in Table 1.1. Pre-Study Material Used with all Participants Background survey (Appendix A) All participants finished a survey aimed at getting background information about their speech and language abilities. Information included the general speech and communication skills and educational background. There were two forms for this survey: one used with the PWS and the other with the control group (i.e., PWNS). Th e survey asked participants about any speech, language, vision, or hearing problems they might exhi bit. The survey also asked the participants

PAGE 42

42 about their education level, type of school they went to (i.e., two year vs. four-year College) in addition to the highest educational degree earned by the participant. WAIS vocabulary The Vocabulary subset of the WAIS test is usually used as a general measure of intelligence and of verbal comprehension. It can al so provide information about the participants education and life experiences (see Zimmerman a nd Woo-Sam, 1973 for a de tailed description of this test). The test consists of 35 words, whic h the participant is asked to define. The words on the test are listed in order of difficulty and ha ve equal numbers of nouns and verbs, plus several adjectives and one adverb. As explained by Zimmerman and Woo-Sam (1973), Vocabulary indicates sensitivity to new information and ideas and the ability to st ore and associatively regroup these as the occa sion demands.(p. 108). The mean score for the PWNS was 62.79 out of 70 and the mean score for the PWS was 54 out of 70. The results of the t tests for the WAIS vocabulary test in dicated a significance difference between the two groups ( t (26) = 2.538, two-tailed p = 0.017). Digits forward and digits backward The tests measure immediate aud itory recall, attention, concentr ation, and the ability of the participant not to be affected by distracters. In ge neral, the point of interest in the tests is to identify whether the participants responses on the test show any discrepancy between digit forward and digit backward recall an d if that is the case, the direc tion of this discrepancy (i.e., in favor of digit forward or digit backward). The mean score on the Digits Forward test for the group was 10.93 out of 14 and the mean score for the PWS was 9.50 out of 14. The results of the t tests indicated no significant difference between the two study groups on the digits forward test ( t (26) = 1.286, two-tailed p = 0.210).

PAGE 43

43 The mean score on the Digits Backward test for the PWNS was 9.07out of 14 and the mean score for the PWS was 7.29 out of 14. The results of the t tests indicated no significant difference between the two study groups ( t (26) = 1.632, two-tailed p = 0.115). Digit ordering This test is used to assess linguistic working memory and/or langua ge processing skills (see Maryellen C. MacDonald, Amit Almor, Victor W. Henderson, Daniel Kempler, and Elaine S. Andersen, 2001 for details). The mean score for the PWNS group was 20.43 out of 24 and the mean score for the PWS group was 17.21 out of 24. The results of the t test s indicted a significant difference between the two study groups on the digit ordering test ( t (26) = 2.5380, two-tailed p =.010). Pre-Study Material Used with PWS Stuttering severity level task (Appendix A) Two speech samples, each of which was approximately 300 syllables long, were elicited from each of the participants in the experimental group. This task was used to provide additional information about the speakers overall level of speech fluency. To elicit the speech samples, each of the PWS was asked to talk for 3 minutes about the plot of a movie they had seen recently, and then for another three minutes about a pleasant experience they had. The resulting speech samples were audio and video recorded an d later analyzed for (a) types of disfluency produced (i.e., repetitions and prolongations such as the prolongation of sounds, or the repetition of sounds or parts of words, the re petition of monosyllabic words, multi-syllabic words, or phrases, or other disfluencies such as revisions of words or phrases and interjections) (b) percentage of syllabl es stuttered in a speech sample of 100 syllables.

PAGE 44

44 Self-rating scale (Appendix A) In addition to the above pre-study tasks, each of the participants who stutter finished a self rating scale intended to get general informati on about each participants stuttering severity during everyday situations. Preparation and Description of the Mate rial Used in Synt actic Priming Task The sentences and pictures used in the present study were originally developed and used in syntactic priming studies conducted by Bock a nd colleagues and later adapted and used by Dr. Lori Altmann at the Language over the Lifesp an Laboratory at the Un iversity of Florida (Bock, 1986, 1989; Bock & Griffin, 2000; Bock, Leobell, & Morey, 1992). The material consisted of priming sentences and associated picture represen tations, plus a variety of filler sentences and their associated picture representations. The rema inder of this section de scribes those sentences and their associated pictures in detail. It is worth mentioning th at although I provide a detailed description of the study material below, the material used was generously shared by Dr. Lori Altmann at the University of Florida who used the same sentences and pictures to elicit responses in studies using methodology similar to the one used in the present study. The study material consisted of 54 sets of pr iming sentences, which were paired with 54 pictures of events. A given primi ng sentence set was classified as transitive, dative or two-clause based on (a) the main verb type in each sentence, (b) the syntactic complexity of each sentence as described below in detail (See Appendix B fo r a list of all the priming sentence sets). The following section provides a detailed descripti on of each of these three sentence types. Transitive priming sentences and pictures There were eighteen sets of transitive primi ng sentences each of which included an active sentence, a corresponding full passive (a by pa ssive) sentence, and a locative sentence. The locative sentences were used in Bock and collea gues studies as a contro l condition to prime the

PAGE 45

45 passive sentences since the two sentence types ha d a similar phrase structure (see below). The structure of the active sentences in the transi tive priming set was as follows: a subject noun phrase, followed by a transitive ve rb and a direct-object noun phrase as in Example (1): (1) The construction work er drove the bulldozer. The structure of the corresponding passive sent ence in the transitive priming set was as follows: subject noun phrase, followed by a full passive verb, and the preposition by followed by a prepositional noun phrase as in Example (2) (2) The bulldozer was driven by the construction worker The structure of the locative sentence in the transitive priming set was as follows: a subject noun phrase, followed by a verb in the pres ent participle form, and the preposition by followed by a prepositional noun phrase as in Example (3): (3) The construction worker was digging by the bulldozer. Each of the transitive priming sets was matche d with one target pict ure intended to elicit transitive verbs (active and passi ve). Each of the transitive verb pictures depicted an action involving an agent and an object undergoing the action (i.e., each tr ansitive picture included two actors in the pictured event). Th e actions depicted in the pict ures included driving, destroying, striking, chasing, kicking, towing, stinging, hi tting, kissing, pushing, an d squirting (e.g., lightning striking a church, and a wrecking ball de stroying a building). The action in each of the transitive pictures could be described using eith er an active transitive or a passive transitive sentence structure. To illustrate, one of the tran sitive sets included the fo llowing three sentences: Active: The lumberjack struck the giant redwood tree; Passive: The lumberjack was struck by the giant redwood tree; Locative: The lumberj ack was eating by the giant redwood tree. The

PAGE 46

46 picture matched with this set was of a fireman rescuing a baby from fire (See Appendix B for a full list of the transitive sentence sets and ta rget picture that was matched with each set). Dative priming sentences and pictures There were eighteen sets of dative verb sent ences each of which included a prepositional dative sentence, a corresponding double-object da tive sentence, and an object complement sentence. The object complement sentences were used in Bock and colleagues studies as a control condition to pr ime the double-object sentences sinc e the two sentence types have a similar phrase structure as described below. The structure of the prepositional dative sentence was as follows: a subject noun phrase, followe d by a dative verb, which was followed by a direct-object noun phrase, and a prepositional noun phrase beginning with to as in Example (4): (4) The children sang a s ong to their babysitter. The structure of the corresponding double-object sentence in th e dative priming set was of a subject noun phrase, followed by a dative ve rb, which was followed by a direct-object noun phrase, and an indirect-object noun phrase, as in Example (5): (5) The choir sang the wedding guests a new hymn. The structure of the control object complement se ntence in the dative priming set was as follows: a subject noun phrase, followed by a dative ver b, which was followed by a direct-object noun phrase, and an object complement as in Example (6): (6) The choir considered the new hymn their favorite song. Each of the sentences in the dative priming se ts was matched with a target picture intended to elicit dative verbs. Each of the dative verb pictures depicted an action involving an agent, an object undergoing the action, and a r ecipient of the action (i.e., each dative picture included three actors in the pictured event). The actions depicted in the pi ctures included giving, showing, throwing, passing, reading, serving, and handing (e .g., a librarian giving a book to a boy, and a

PAGE 47

47 waitress serving drinks to a man). Each of the da tive priming pictures co uld be described using either a double-object dative or a prepositional-dative sentence structure. To illustrate, one of the dative sets included the three foll owing sentences: pre positional dative The diplomat took the secret documents to the president, double-object dative The diplomat took the president the secret documents, and object complement The diplomat made delivering the secret documents his primary mission. The picture ma tched with this set was of a girl and a boy showing a picture to a teacher (See Appendix B for a full list of the dative sentence sets and ta rget picture that was matched with each set). Two-clause priming sentences and pictures There were eighteen sets of two-clause se ntences each of which included a conjoined sentence (i.e., two independent clauses joined using the coordinating conjunction and, or the subordinating conjunction because), a center-emb edded (subject-subject) relative clause, and a right-embedded (object-object) relative clause (i.e., one independent clau se and one center or right embedded relative clause) as in Examples (7), (8), (9) respectively: (7) The duck is kicking the girl and the girl is kicking the boy. (8) The girl that is kicked by the duck is kicking the boy. (9) The duck is kicking the gi rl that is kicking the boy. Each sentence in the two-clause priming senten ce sets was coupled with a picture that was intended to elicit sentences that can be descri bed using either a conj oined clause, a centerembedded clause, or a right-embedded clause. To illustrate, one of the two-clause sets included the three following sentences: conjoined The girl smelled the flower and it reminded her of her grandmother, center-embedded The girl that sm elled the flower was very pretty, and rightembedded The girl smelled the fl ower that was very pretty. The picture matched with this set

PAGE 48

48 was of a man pulling a woman who was pulling a dog (See Appendix B for a full list of the dative sentence sets and target pictur e that was matched with each set). Filler sentences and pictures In addition to the priming sentences and targ et pictures described above, there were 127 filler sentences and pictures (See Appendix B for a list of the filler sentences). The filler pictures were similar in style and preparation to the expe rimental pictures but de picted actions that are commonly described with sentence forms differen t from those used in the priming sentences. Forms used in the filler sentences included refl exives, and predicate adjective sentences as in Examples (10) and (11) respectively: (10) The surgeon cut himself with a scalpel (11) The books were expensive. Using the above described senten ces and pictures, three lists were developed and used in the study. Each of the lists contained 252 priming tr ials that consisted of random presentations of a priming sentence, priming picture, filler senten ce, or filler picture. Each of the priming sentences was immediately followed by a priming pict ure. Thus a typical sequence of stimuli in a priming trial was: filler sentence, filler picture, priming sentence, priming picture. The filler sentences were randomly interspersed within each list and could either have the sequence of one, two, or three consecutive filler sentences followe d by one, two, or three filler pictures. There were ten filler sentences between each two priming sentences. The transitive, dative, conjoined, and embedded priming sentence sets alternated in each list so that an equal number of each sentence struct ure occurred in each list (6 sentences each of active transitive, passive transitive, locative, double-object dative, prepositional dative, object compliment, conjoined, center-embedded, and right-e mbedded, resulting in 9 sentence types x 6 instances of each type, which equaled a total of 54 priming sentences and 54 corresponding

PAGE 49

49 target pictures in each study list). The different se ntence structures alternated in the lists in a way that ensured participants did not encounter two priming sentences or tw o target pictures of the same type in successive priming trials. Each li st included only one sentence type from each set. For example, list A included the transitive activ e sentence The lumberjack struck the giant redwood tree, list B included th e corresponding passive sentence fr om that set The lumberjack was struck by the giant redwood tree and list C included the locative senten ce of that set The lumberjack was sitting by the giant redwood tr ee. The same procedure was followed for the distribution of all other sentence sets among the three study lists. Th e order of the priming trials was the same in the three lists. Following the data collection session, each participants audio and video-recorded speech sample was transcribed orthograp hically and then coded to get specific information for each sentence. Information about sentence type, sp eech reaction time, and fluency for sentences produced by each of the participants wa s determined as described below under Data Analysis. Apparatus A digital voice recorder (Olym pus WS-100) was used to reco rd responses during the pre and main study tasks. During the experiment, the stimuli were presented on a laptop computer (Dell Inspiron 15150) and the speech productions were recorded using a lightweight headset microphone (Optimus 33-3012) that was approxima tely two inches from the speakers mouth. The study material was presented using Direct-RT (DirectRT 2006 2.16) a graphical experiment generator which is capable of managing and anal yzing data during time controlled experiments such as the one used in the present investigation Direct-RT also recorded in milliseconds each participants SRT for picture de scription during the study task.

PAGE 50

50 Procedures Data Collection Each of the percipients was s een individually for one session divided into two parts. Upon successful completion of eligibilit y testing in the first part of the session (described above under Material), participants finished the second part of the session (i.e., the syntactic priming task) during which SRT and fluency data collection took place. The first part of the session took 20 to 25 minutes to complete, and the second part t ook 45 to 60 minutes to complete. It is worth mentioning that the participants finished a seco nd task (i.e., sentence generation task); however, data and results for responses during that task ar e not reported in this paper. The following is a detailed description of the procedures followed during the syntactic priming task. I described the task to the part icipants and told them that the first 9 items were practice material. The participant saw either a picture on th e computer screen or the sentence, Listen and Repeat. When they saw a picture, they were in structed to make up a one-sentence description of the event pictured and say it aloud as soon as possible after the pictur e appeared. They were also instructed not to use pronouns in their pictur e. This instruction was added to motivate the participants to produce the appr opriate number of noun phrases representing the main actors in the depicted events such as The bear is kicking the girl and the gi rl is kicking the boy instead of a partial description such as They are kicki ng each other when a pronoun is used in that response. When the participants saw the sentence Listen and Repeat, I read a sentence and the participants were instructed to repeat the sentence I just read. The participants were told that the experimenter could read the sentence ag ain if it was not clear the first time. If the participant responded successfully to th e first 9 items on the list, the experimental sentences and pictures were presented. In instan ces were participants response to a trial item was not accurate or when the participant seemed confused about the nature of the expected

PAGE 51

51 response, I explained the error made by the partic ipant on that trial and as ked the participant to repeat his/her response to that tr ial following the correct instruc tions. After finishing the first 9 items in the task, the main task items were presented individually on the computer screen. For each trial, the instruction Listen and Repeat and the picture remained on the screen until the participant repeated the sentence or fini shed the description of the pictured event. Immediately after each response, the participants were asked if theyve heard the sentences or seen the pictures before within the course of the experiment by answer ing the questions Have you heard this sentence before? that immediat ely followed their sentence repetition and the question Have you seen this picture before? which immediately followed their picture description. Participants answered the question by responding yes or no. The questions were used as part of a recognition memory test following Bock (1986) procedures to minimize subjects attention to their spee ch and its structural features (p. 360). The question remained on the screen until the participant responded to that question. I then pushed a mouse button which in turn triggered the software program to move to the next trial. The participants were told when the practice ite ms were presented that the task is rather long (on average took participants about 50 minutes to finish) and that they would be offered several opportunities to rest, if n ecessary, during the task. Participan ts were also told that they could ask for a break at any time during the task. Data Preparation for Analyses To summarize the method used to generate th e data for this study, I transcribed verbatim all sentences produced by the participants. Each se ntence was then coded as a target sentence or filler sentence. All filler sentences were excluded from further analyses. The remaining sentences were further coded for general desc riptive characteristics (i.e., nu mber of syllables and words in each sentence), grammaticality (i.e., whether th e sentence was grammatical or not), sentence

PAGE 52

52 type (whether the sentence was transitive, dative, two-clause, or other), and for fluency measures (i.e., whether the sentence (a) was all fluent, (b) contained a speech disfluen cy in any part of the sentence, (c) contained a speech disfluency in the first noun phrase in the sentence (d) contained a speech disfluency in the first word of the first noun phrase in the sentence; as well as the number of disfluencies in each sentence, and th e type of disfluencies in each sentence). The transcripts were used to determine re sponse fluency (i.e., fluent or disfluent; repetitions and prolongations, or other types of disfluencies, and the location of disfluency within an utterance) and respons e structure (i.e., Transitive: activ e, passive; Dative: prepositional dative, double-object dative; Two-clause: conjoi ned, embedded). The following are the details of data preparation for analysis. Data excluded from the final analyses Three main data analyses were of interest in the present study: flue ncy analysis, sentence type analysis, and SRT analysis. For each analys is, several responses were excluded from the final analyses based on criteria fo llowing either procedures used in other studies (e.g., Bock et al. 1986, Logan, 2003). Out of 1620 possible responses for the two groups (i.e., 54 responses x 30 participants), 1617 sentences were actually produ ced and a total of 328 (20.28%) sentences were excluded from the final analyses based on the app lication of a number of exclusion criteria which are described below. The first of the sentence exclusion criteria concerned elicitation and response problems (i.e., problematic responses). A problematic re sponse was defined as one that was associated with (a) an unrelated vocal respon se preceding the onset of the target verbal response (coded as a false start e.g., a yawn, cough, or laugh by the pa rticipant, and responses that started with a question or a comment by the participant that was not part of the target response), (b) program error (i.e., the software program used to presen t the priming pictures skipped an item), (c)

PAGE 53

53 presentation of the wrong priming sentence by the experimenter (i.e., I read the wrong priming sentence or one that was not asso ciated in its syntactic structure to the target picture). The total number of problematic responses was 107 (6.61 %) responses (64 false starts, 32 program skips, and 11 wrong sentences). Thus, following applica tion of the first of the sentence exclusion criteria, 1510 sentences remained for analysis. The second of the sentence exclusion criter ia concerned response grammaticality. An ungrammatical response was defi ned as one that was missing a main verb or parts or a verb (e.g., A boy running,) or exhibited inaccurate word form or word choice (e.g., The bulldozer is demoralizing the building, The pedestrian got hit by a rescue,, and The turtle wet the cat). The total number of ungrammatical respons es was 112 sentences (6.92 % of 1511 remaining responses). Thus, following application of the second of the exclusion criteria, 1398 sentences remained for analysis. The third of the sentence ex clusion criteria concerned exam ination of speech reaction time for each sentence produced by the participants Recall that SRT in milliseconds for each participant was captured by the computer and was defined as the time from the onset of the presentation of the priming picture on the comput er screen to the onset of verbal response by the participant captured by the voice activated micr ophone. To eliminate the effect of outlier SRT values, and after consultations with two co mmittee members, I trimmed the SRT data by including only the SRT values that fell between 500 milliseconds, at the low end, and three standard deviations above the SRT mean value for a given participant, at the high end. Thus, all SRT values for grammatical sentences below 500 milliseconds as well as SRT values that fell 3 standard deviations above the mean SRT value fo r a participant were excluded from subsequent analyses. The total number of excluded SR T outliers was 63 sentences (3.89 % of 1398

PAGE 54

54 remaining responses). Following the application of the third exclusion criteria 1335 sentences remained for analysis. The fourth and last of the ex clusion criteria concerned overa ll participant performance. One participant in the experimental group produc ed only two grammatical sentences out of the 54 possible responses thus resulti ng in only 2 usable sentences. As a result, responses for that participant were excluded from the final statistical analyses, as were response for the matched participant from the control group. Thus, in all, a total of 46 possible responses (produced by those two participants) or 2.84 % out of 1335 remaining responses were excluded from consideration. Accordingly, statistic al analyses described in the subsequent sections are based upon sentences produced by the remaining 28 participants (14 PW NS and 14 PWS). After all the sentence exclusi on criteria were applied, the da ta file used to conduct all subsequent analyses included only grammatical sentences, that we re (a) entirely accurate (i.e., they contained no false starts or evidence of pr ogram or experimenter errors) and (b) free from outlying SRT scores. Thus, 1289 responses were in cluded in the final an alyses. This number represented 79.71 % of the total se t of responses that the participants originally produced. General descriptive analyses Each participant had the opportunity to pr oduce 252 responses: 108 cri tical trials (i.e. repetitions of 54 priming sentences and generation of 54 sentences to descri be target pictures), and 144 fillers (i.e., repetitions of 72 filler sentences and generation of sentences to describe each of the 72 filler pictures). Only pict ure descriptions during the critical trials were of interest in this study, thus the sentence fluency an alyses were based on the 54 possible sentences generated by each participant to describe the priming pictures. All other responses (i.e., repetitions of priming and filler sentences and descriptions of filler pictures) were excluded from the final analyses.

PAGE 55

55 Data preparation for speech fluency analyses Each sentence produced by the participants wa s coded as to whether it was fluent or disfluent. Fluent sentences were those that di d not contain repetitions and prolongations as defined in (1) to (7) below a nd other disfluency types as defined in (8) and (9) below: 1 Sound or syllable repetition: Instances at which the part icipant repeated a sound or syllable in a given word as in The b boy is giving the girl a valentines card, and The bear is hiding behind the tree while the boy is look looking at the duck. 2 Monosyllabic word repetition: Instances at wh ich the participant repeated a one-syllable word as in The the man is cleaning the window. 3 Polysyllabic word repetition: Instances at wh ich the participant repeated a word that contained more than one syllable as in The hostess is serving serving drinks to her guests. 4 Phrase repetition: Instances at which the par ticipant repeated a string of words within a sentence as in The bee the bee is stinging a man. 5 Inaudible sound prolongation: Instances at whic h the participant seemed to start a word and block or exhibit difficulty moving from the initial sound or part of that word to the next sound as in The p policeman is handing the man a ticket. 6 Audible sound prolongations: In stances at which the part icipant started a sound and vocalization of the sound could be heard as in The fff firefighter is rescuing a baby from the fire. 7 Broken words: Instances at which participants produced a sound or syllable repetition or an inaudible sound prolongation at syllable divisions within a word as in The construction worker drove the bull do dozer, and The boy is hold ding a girls hand. 8 Other disfluency types: Interjections and re visions as in Um, The construction worker drove the uh bulldozer. 9 Revisions: Instances at which the particip ant responded with a phr ase or a sentence, stopped, and changed their choice of that phrase or sentence as in A clier, a fireman is climbing out of a building and A girl is squirting a boy with a gun, with a squirt gun, and A man and a woman, oh no, A man is l ooking at a baby and a woman is looking at the man. Disfluent sentences produced by each group were further coded in terms of the position of the disfluency (e.g., 1st word, 1st noun phrase, etc.) within the response. This analysis was performed to be used in providing general descrip tive data for fluency and was also used to guide

PAGE 56

56 the manner in which statistical analyses for the SR T analyses were to be conducted (described in more detail below). Sentences that were produce d with no disfluencies we re coded as fluent. Overall, participants produced 646 entirel y fluent sentences (50.11 %) and 643 disfluent sentences (49.88 %). In terms of group descrip tive statistics, 76.22 % of sentences produced by PWNS were entirely fluent and 23.92% were disflu ent. As expected, PWS showed a pattern that was nearly opposite the one that was observed fo r PWNS, that is, only 21.59% of the sentences produced by PWS were entirely flue nt and 78.40 % were disfluent. Data preparation for sentence type analyses Following data preparation procedures simila r to those used for the fluency analyses, sentence type analyses was based on the same 54 sentences generated by each participant to describe the priming pictures. All other responses (i.e., repetitions of priming and filler sentences and descriptions of filler pictures) were excluded from further analysis. Each of the 54 sentences produced by a participant was assi gned to one of the following cat egories: Transitive as in examples (1) and (2) above under Transitive prim ing sentences and pictures, Dative as in examples (4) and (5) above under Dative priming sentences and pictures, and Two-clause as in examples (7), (8), and (9) above under Two -clause priming sentences and pictures. If a sentence was coded as Transitive, it was further coded as either active or passive. To be coded as an active, the sentence had to have an acceptable corresponding passive sentence form. A passive form had to include a passive verb (i.e., a main verb proceeded by a form be or get and followed by a by-phrase (i.e., prepositional phrase starting with the preposition by). For example, an active sentence was The dog is ch asing a mailman and the corresponding passive sentence was The mailman is chased by the dog (see Transitive priming sentences and pictures above for detailed desc ription and examples of the activ e and passive sentence types).

PAGE 57

57 Passive sentences that did not include the by-phra se were coded as other (e.g., A man is bitten in the arm). If a sentence was coded as Dative, it was furthe r coded as either pre positional dative, or double-object dative. To be coded as Dativ e, the sentence had to have an acceptable corresponding double-object sentence form (see D ative priming sentences and pictures above for detailed description and examples of the prepositional and double-object sentence types). Dative sentences had to include appropriate num bers of noun phrases indicating the three main actors in the depicted action (e.g., The girl (1) is giving the man (2) a paint brush (3)). Dative sentences that did not include a ll three main actors (e.g., The girl(1) read to the boy(2), as compared to The girl(1), read a story(2), to the boy(3)) were coded as other. An additional category: dative + two-clause wa s added to the Dative senten ce type analysis. The dative + two-clause category was added to show that in some instances, the Dative and Two-clause categories were not mutually exclusive. If a se ntence was coded as two-clause, it was further coded as either conjoined or embedded. (s ee Two-clause priming sentences and pictures above for detailed description and examples of the conjoined and emb edded sentence types). Finally, other sentences structur es (i.e., non-transitive, reflexiv es, and predica tive adjective sentences as in examples (10), (11), and (12) respectively above under Fill er sentences pictures were coded as other. The number of sentences produced per each examined sentence type is illustrated in Figure 2-1. Pertaining to response length for each of the ex amined sentence types a nd as illustrated in Figure 2-2., on average, and for participants in the two study groups, transitive sentences were shortest in terms of both words (PWNS = 7.97, PWS = 7.74) and syllables (PWNS = 10.26, PWS

PAGE 58

58 = 9.96) number, and two-clause sentences were longest (words: PWNS = 12.60, 12.63; syllables: PWNS = 15.43, PWS = 1620). Data preparation for SRT analyses SRTs for each of the sentences were code d online by the Direct-RT program. SRT was measured (in milliseconds) and defined as the time of picture (target or filler) presentation to the time of acoustic onset for participants response. Recall that du ring picture description trials for both the target and filler pictures, the Direct-RT program generated and controlled the time at which pictures were presented. The program also r ecorded the latencies of the participants vocal response, in milliseconds, for each of the pictures. The overall mean SRT for PWNS was 2667.20 ms (S.D. 1252.28 ms) and for the PWS, was 2596.78 ms (S.D. 1532.33 ms). Based on thes e values, SRT for PWS was on average 70.42 ms shorter than SRT for PWNS. In terms of data preparation for the SRT analyses per senten ce type, it is worth mentioning that deciding on which responses to include in the SRT analyses was not entirely obvious. The difficulty in making this decision lies in the fact that when earlier studies used sentence structure priming methodology to examine structure priming effects in addition to SRT data, almost all these studies used a very rigid de finition of what a fluent response is. For example, in Bocks studies (e.g., Bock, 1986; Bock 1989) that I reviewed, whenever a reference was made to what a fluent response is, the definiti on stated that such a response is one that was produced with no entirely fluent (i.e., exhibited no speech disfluencies such as in terjections, revisions, hesitations, or sound or syllable repetitions). Note however that none of the above-referenced studies by Bock examined SRT and fluency in PWS. As w ould be expected based on previous research, PWS on average produce significantly more disflu ent responses than PWNS. As a result, when comparing sentence production between these tw o groups, and when participants in the two

PAGE 59

59 groups are given equal opportunitie s to produce a similar number of responses, a percentage of those responses will be disfluent; however, a substantial number of the disfluent responses will be produced by the stuttering group. If such di sfluent responses are excluded from a given analysis, this will in the end result in the loss of several data points for participants in both groups but substantially more for the participants in the stutte ring group. In this study, out of 673 responses produced by PWNS, 513 (76.52%) were en tirely fluent compared to 616 responses produced by PWS out of which only 133 (20.25%) were entirely fluent. Cl early, if this rigid definition of fluency is used in the present study, it would be necessary to exclude at least 4 of the 14 PWS and the sample means for many other participants would be based on fewer than 10 data points. For example, looking at the range for how many fluent transitive sentences were produced by each group, PWNS on average produced 13 fluent transitive sentences (range 6-18). PWS on the other hand produced on average about 4 fl uent transitive sentences (range 0-10). The averages for the number of fluent dative and two-clause sentences show a similar pattern. Compare this to when a more lenient definitio n of a fluent response is used. In this case, a fluent response is defined as one that exhibits no disfluencies within the first noun phrase in a sentence (NP1F). As illustrated in Figure 2-3, when using this definition, the number of responses that can be included in the analyses increases for th e two groups, and almost doubles for PWS (547 responses when the sentences exhi biting a fluent first noun phrase were included in the analysis compared to only 133 when only en tirely fluent responses were included). Using the number of transitive sentences again as an example, and when looking at the range for how many transitive sentences exhibiting a flue nt first noun phrase were produced by each group, although the increase is not that dramatic for PWNS who on average produced 13 transitive sentences with a fluent first noun phrase, the percen tage of sentences that can be included almost

PAGE 60

60 doubles (40.93%) for PWS compared to when only entirely fluent sentences were examined (20.25%). When yet, a broader definition is used for se ntence inclusion (i.e., us ing all sentences that exhibited a fluent first word in the first noun phr ase of a sentence), the percentage of sentences that can be included for PWS yet increases ma rkedly (57.38%; on average, 9 sentences per PWS). Thus, based on the above, and to objectively represent the results of any effects or interactions among the different va riables in the present study, in addition to including as many responses as possible in the anal yses, the reported SRT analyses and results are for sentences in which the first noun phrase in the sentence was fluent (NP1F). It was felt that such an approach would offer a reasonable compromise between bei ng forced to use mean SRT values that were based on relatively few data points and using mean SRT values that might be conflated with sentence factors having to do with stuttering within the subject constituent of the first NP. Intrajudge and Interjudge Measurement Reliability All fluency and sentence type measures in this study were made by the author. Inter and intrajudge measurement reliability for sentence type were conducted by randomly selecting two participants and reanalyzing al l the sentences produced by thos e two participants (total 108 sentences or 7 % of the data). I first recoded the 108 sentences fo r sentence type coding and then a graduate student who had a background in lingu istics and who was familiar with the sentence type coding processes used in the study independently recoded each sentence for sentence type. Intrajudge analysis showed 99% agreement (disag reement was a coding error for one sentence), and interjudge analysis show ed 98% agreement (disagreemen t was a coding error for two sentences). The interjudge points of disagreement were discussed and resolved.

PAGE 61

61 Inter and intrajudge measurement reliability for fluency were conducted by randomly selecting two participants and reanalyzing all the sentences pr oduced by those two participants (total 108 sentences or 7 % of the data). I fi rst recoded the 108 responses for fluency (i.e., fluent or disfluent) and then recoded the di sfluent responses for type of disfluency (i.e., repetitions and prolongations, other) and posi tion of disfluencies within response (i.e., 1st word, 1st noun phrase, etc.). Then an undergradu ate student who was fami liar with the fluency coding processes used in the study independently recoded each response for fluency. Intrajudge analysis showed 98% agreement, and interjudge analysis showed 94% agreement. The points of disagreement were discussed and resolved.

PAGE 62

62 Number of Responses per Sentence Type 0 5 10 15 20 25 TransitiveDativeTwoClause Other Sentence TypeMean number of responses PWNS PWS Figure 2-1. The mean and standa rd deviation for the number of responses produced by persons who do not stutter (PWNS; N = 14) and pers ons who stutter (PWS; N = 14) across the examined sentence types (Transitive, Da tive, and Two-Clause) in addition to responses that exhibited synt actic structures other than the examined types (Other). Response Length in Words 5 10 15 TransitiveDativeTwo-Clause Response TypeWords Per Response PWNS PWS Figure 2-2 Mean and standard deviation for number of words produced by persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) for (a) Transitive responses in the Transitive experiment (b) Dative responses in the Dative experiment, and (c) Two-Clause respons es in the Two-Clause experiment.

PAGE 63

63 Percentage of Usable Responses Based on Fluency Definition 0 20 40 60 80 100 Entirely FluentNP1 FluentWord1 Fluent GroupPercentage of usable responses per fluenc y definition PWNS PWS Figure 2-3. The mean and sta ndard deviation for the percen tage of sentences produced by persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when the sentence was entirely fluent, when the first noun phrase in the sentence was fluent, and when the first word in the fi rst noun phrase in the sentence was fluent

PAGE 64

64Table 2-1. Participant Demographics and Perfor mance on the Pre-study Language and Memory Tests ID Number Age Gender Years of Education Digits Forward Max = 14 Digits Backward Max = 14 Digit Ordering Max = 24 WAISVocabulary Max = 70 Stuttering Severity Level Stuttering Treatment S02 40 F 14 5 5 17 54 Severe No S03 21 F 15 7 5 12 62 Moderate-Severe Yes S04 34.8 F 16 10 7 14 60 Moderate Yes S05 24.1 M 17 12 7 21 50 Severe Yes S06 18.2 M 12 14 14 22 61 Mild Yes S07 19 M 13 12 6 20 56 Severe Yes S08 54.1 M 17 7 6 20 60 Moderate-Severe Yes S09 29.4 M 17 11 12 17 66 Moderate Yes S10 44.7 M 17 10 9 21 68 Moderate-Severe Yes S11 56 M 12 13 4 16 40 severe Yes S12 18 M 13 5 8 17 38 Moderate Yes S13 27 M 17 7 6 12 34 Moderate Yes S14 44.1 F 19 6 4 15 43 Moderate Yes S15 59.8 M 19 14 9 17 64 Severe Yes NS01 21.4 M 17 10 6 23 58 NA NA NS02 22.1 M 17 8 6 18 58 NA NA NS03 34 F 20 14 13 21 69 NA NA NS04 21 F 17 8 8 21 48 NA NA NS05 27.2 M 21 12 13 24 62 NA NA NS06 21.8 M 16 11 8 22 65 NA NA NS07 24 M 17 8 8 18 62 NA NA NS08 26.8 M 20 13 12 22 69 NA NA NS09 43.1 M 18 14 8 21 69 NA NA NS10 59.3 M 21 12 12 16 64 NA NA NS11 34.9 F 29 13 10 23 69 NA NA NS12 53 M 14 6 6 14 53 NA NA NS13 43 F 14 11 5 21 69 NA NA NS14 24 M 17 13 12 22 64 NA NA

PAGE 65

65 CHAPTER 3 RESULTS Recall that in this study, I was interested in examining the differences between adults who stutter and adults who do not stutter in speech fluency, speech reaction time, and responses to priming across sentences varying in their syntactic structure. Seve ral analyses were conducted to examine differences between the two groups in (a ) the overall frequency of disfluencies they produced, (b) the frequency of disfluencies produce d pertaining to sentences of varying syntactic structure, (c) the effect of prim ing on the frequency of disfluencies they produced, (d) the extent to which syntactically primed structures were incorporated into res ponses, (e) overall mean SRTs, and (f) mean SRT for each of several sentence types. This chapter on results is organized into th ree main subsections that are based on the analyses performed to address the above points. The first subsection deals with the fluency analyses. In this section, results pertaining to the statistical anal yses that address points (a) and (b) above are described. The sec ond subsection deals with syntactic priming analyses. In this section, results pertaining to the statistical anal yses used to address poi nts (c) and (d) above are described. The third subsection deals with the sp eech reaction time analyses. In this section, results pertaining to the statistica l analyses used to address points (e) and (f) above are described. Fluency-Related Results Three analyses were used to answer the flue ncy-related questions. Th e analyses were based on the three sentence type experiments in the present study (i.e., Transi tive, Dative, and Twoclause). The first analysis examined the frequenc y and type of speech disfluencies produced by each group during the Transitive experiment acro ss the different prime types (i.e., active, locative, and passive primes). The second analysis examined the frequency and type of speech disfluencies produced by each group during the Dative experiment across prime types (i.e.,

PAGE 66

66 double-object, object-complement, a nd prepositional object primes). The third analysis examined the frequency and type of speech disfluencies produced by each group during the Two-Clause experiment across the different prime types (i.e., conjoined, center -embedded, and rightembedded primes). A fourth analysis was conduc ted to compare the frequency and type of speech disfluencies produced by each group acro ss the three experiments (i.e., Transitive, Dative, and Two-Clause). In the remainder of th is section, results of these four analyses are presented. Transitive Experiment Figure 3-1a shows the frequency with whic h repetitions and prol ongations and other types of disfluencies were produced by partic ipants in the two groups in the Transitive experiment. As can be seen, overall, the PW S seemed to produce more repetitions and prolongations across the three examined transitive conditions (i.e., active, locative, passive) than the PWNS. To examine fluency in the Transitive experi ment, a 2 (Group) 2 (Disfluency Type) 3 (Prime Type) multivariate analysis of variance (MANOVA) was conducted, with Group (PWNS, PWS) as the between-subjects factor and Disf luency Type (repetitions and prolongations, other) and Prime Type (Active, Locative, Passi ve) as within-subjects factors. The dependent variable was the frequency of repetitions and prolongations and other types of disfluencies the participants produced per Transitive prime type. The main effect for the within-subjects factor Disfluency Type was significant ( F (1, 26) = 16.562, p < 0.001, 2 = 0.389) and the Group Disfluency T ype interaction was also significant ( F (1, 26) = 24.143, p < 0.001 2 = 0.481) indicating that the di fference between the frequency of repetitions and prolongations and the frequency of other disfluencies per response for one study group were different from t hose frequencies for the other study group. The main effect of

PAGE 67

67 the between-subjects factor, Group was also significant ( F (1,26) = 34.50, p 0.001, = 0.57) indicating that the PWS produced significantly more overall speech di sfluencies in the transitive responses ( M = 0.80) than the PWNS ( M = 0.11). To further examine the Group Disfluency interaction, posthoc paired-samples t tests were run comparing the within-s ubjects factor (i.e., frequency of repetitions and prolongations and other types of disfluenci es) in the following experiment al condition pairs to determine which were significantly different: Repetitions and prolongations produced by PWNS versus r epetitions and prolongations produced by PWS Other disfluencies produced by PWNS ve rsus other disfluencies produced by PWS A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/2 t tests = .025). The results of one comparison was signi ficant indicating that the PWNS produced significantly fewer repet itions and prolongations per response following the transitive primes examined in the present study ( M = 0.067) than PWS ( M = 1.28) ( t (26) = -5.864, two-tailed p < 0.001). The results of the other co mparison was not significance ( t (26) = -1.898, two-tailed p = 0.069) indicating that although PWNS produced fewe r other types of disfluencies in the transitive responses following the transitive prim es examined in the present study than PWS, the difference in the frequency of other types of disfluencies was not sign ificant between the two groups. None of the results of the Prime Type anal yses for the transitive experiment were significant, indicating that for all participants, the transitive prime types examined in the present study did not seem to affect the frequenc y of repetitions and prolongations versus other disfluency types produced in responses to those primes.

PAGE 68

68 In summary, the results of the Transitiv e experiment analyses indicated that PWS produced significantly more repetitions and prolongations than PWNS. The results also indicated that particip ants in the two study groups produced a comparable number of other disfluencies in responses following the transiti ve primes examined in the present study. In addition, the results indicated that for all part icipants, the frequency and type of speech disfluencies produced in the transitive responses did not seem to be affected by whether or not the priming sentences preceding those responses exhibited an active, locative, or passive transitive sentence structure. Dative Experiment Figure 3-1b shows the frequency with which repetitions and prol ongations and other types of disfluencies were produced by participan ts in the two groups in the Dative experiment. As can be seen, overall, the PWS seemed to produce more repetitions and prolongations across the three examined dative conditions (i.e., doub le-object, object-complement, prepositionalobject) than the PWNS. To examine fluency in the Dative experime nt, a 2 (Group) 2 (Disfluency Type) 3 (Prime Type) MANOVA was conducted, with Gr oup (PWNS, PWS) as the betweensubjects factor and Disfluency Type (repetitions and prolongations, other) and Prime Type (Doubleobject, Object-complement, Prepos itional object) as the within-s ubjects factors. The dependent variable was the frequency of repetitions and prolongations and other types of disfluencies the participants produced per Dative prime type. Consistent with results from the Transitive experiment, the main effect for the withinsubjects factor, Disfluency Type, was significant ( F (1, 26) = 19.912, p < 0.001, 2 = 0.434) and the Group Disfluency Type inte raction was also significant ( F (1, 26) = 21.638, p < 0.001 2 = 0.454), indicating that the differe nce between the frequency of repetitions and prolongations

PAGE 69

69 and the frequency of other di sfluencies per respons e type for one study group were different from those frequencies for the other study group. Th e main effect of the between-subjects factor, Group was also significant ( F (1, 26) = 32.863, p < 0.001, = 0.55) indicati ng that the PWS produced significantly more overall speech di sfluencies for their dative responses ( M = 0.90) than the PWNS ( M = 0.11). To further examine the Group Disfluency interaction, posthoc paired-samples t tests were run comparing the within-s ubjects factor (i.e., frequency of repetitions and prolongations and other types of disfluenci es) in the following experiment al condition pairs to determine which were significantly different: Repetitions and prolongations produced by PWNS versus r epetitions and prolongations produced by PWS Other disfluencies produced by PWNS ve rsus other disfluencies produced by PWS A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/2 t tests = .025). The results of one comparison was signi ficant indicating that the PWNS produced significantly fewer repet itions and prolongations ( M = 0.100) per dative primes than PWS ( M = 1.54) ( t (26) = -5.577, two-tailed p < 0.001). The results of th e other comparison was not significant indicated that following the dative prim es examined in the present study, participants in the two study groups produced a comparable number of other types of disfluencies per dative response (PWNS M = 0.12, PWS M = 0.26) ( t (26) = -2.017, two-tailed p = 0.065). None of the results of the Prime Type analys es were significant indicating that for all participants, the dative prime types examined in the present study did not seem to affect the frequency of repetitions and prolongations versus other disfluency types produced in responses to those primes.

PAGE 70

70 In summary, the results of the Dative experi ment analyses indicated that PWS produced significantly more repet itions and prolongations than PWNS. The results also indicated that participants in the two study groups produced a comparable nu mber of other types of disfluencies. Additionally, the results indicated that for all participants, th e frequency and type of speech disfluencies produced in the dative respon ses did not seem to be affected by whether or not the priming sentences preceding those re sponses exhibited a double-object, objectcomplement, or prepositional obj ect dative sentence structure. Two-Clause Experiment Figure 3-1c shows the frequency with whic h repetitions and prol ongations and other types of disfluencies were produced by partic ipants in the two groups in the Two-Clause experiment. As can be seen, overall, and similar to the results of the two previous analyses, overall, the PWS seemed to produce more re petitions and prolongations across the three examined two-clause conditions (i.e., conjoi ned, center-embedded, right-embedded) than the PWNS. To examine fluency in the Two-Clause experi ment, a 2 (Group) 2 (Disfluency Type) 2 (Prime Type) MANOVA was conducted, with Gr oup (PWNS, PWS) as a between-subjects factor and Disfluency Type (repetitions and prolongations, other) and Prime Type (Right embedded, Center embedded, and Conjoined) as the within-subjects factors. The dependent variable was the frequency of repetitions and prolongations and other types of disfluencies per response type. Similar to the results of the Transitive and Dative experiments analyses, the main effect for the within-subjects factor Disf luency Type was significant ( F (1, 26) = 34.809, p < 0.001 2 = 0.572) and the Group Disfluency Type interaction was also significant ( F (1, 26) = 31.603, p < 0.001 2 = 0.549) indicating, once again, that the difference between the frequency of

PAGE 71

71 repetitions and prolongations and the frequency of other disflu encies per response type for one study group were different from those freque ncies for the other study group. The main effect of the between-subjects factor Group was also significant ( F (1, 26) = 32.86, p < 0.001, = 0.55) indicating that the PWS produced significantly more overall speech di sfluencies their twoclause responses ( M = 0.87) than the PWNS ( M = 0.20). To further examine the Group Disfluency T ype interaction, post-hoc paired-samples t tests were run comparing the within-subjects factor (i.e., frequency of repetitions and prolongations and other types of disfluencies) in the following e xperimental condition pairs to determine which were significantly different: Repetitions and prolongations produced by PWNS versus r epetitions and prolongations produced by PWS Other disfluencies produced by PWNS ve rsus other disfluencies produced by PWS A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/2 t tests = .025). Similar to the results of the two previous analyses, the results of one comparison was significant indicated that the PWNS produced significantly fewe r repetitions and prolongations per two-clause responses exam ined in the present study ( M = 0.21) than PWS ( M = 1.46) ( t (26) = -6.249, two-tailed p < 0.001) and that the number of other types of disfluencies per response was similar between the two groups. As with the two other experime nts, none of the results for th e Prime Type analyses were significant indicating that for all participants, the two-clause prime types examined in the present study did not seem to affect the frequency of repetitions and prolongations versus other disfluency types produced in responses to those primes. In summary the results of the Two-Clause experiment analyses indicated that PWS produced more repetitions and prolongations th an PWNS. The results also indicated that

PAGE 72

72 participants in the two study groups produced a comparable nu mber of other types of disfluency in responses following the two-clause primes examined in the present study. In addition, the results indicated that for all part icipants, the frequency and type of speech disfluencies produced in the twoclause responses did not seem to be affected by whether or not the priming sentences preceding those responses exhibited a right em bedded, center embedded, or conjoined two-clause sentence structure. Fluency Across Picture Types Before describing this analysis in detail, it is worth menti oning that although the present analysis and three above analyses may seem similar, the present analysis examined the overall effect of picture type (i.e., whether a picture was designed to elicit a Transitive, Dative, or TwoClause sentence) on the frequency and type of di sfluencies between the two groups. In contrast, the three previous analyses examined the effect of a given prime type (e.g., Transitive: Active, Locative, Passive) on the frequency and type of speech disfluencies within that experiment. Figure 3-2 shows the frequency with which repetitions and prolongations and other types of disfluencies were produced by participan ts in the two groups in responses to different picture types across the three expe riments (i.e., Transitive, Dative, and Two-Clause). As can be seen, overall, the PWS produced somewhat more repetitions a nd prolongations ( M = 1.47) and more than the PWNS ( M = 0.12) across the three ex amined picture types. To examine participants fluency across the pict ure types used in the three experiments, a 2 (Group) 2 (Disfluency Type) 3 (Pictu re Type) MANOVA was conducted, with Group (PWNS, PWS) as a between-subjects factor and Disfluency Type (repetitions and prolongations, other) and Picture Type (Trans itive, Dative, Two-clau se) as within-subjects factors. The dependent variable was the frequency of each type of disfluency per trial.

PAGE 73

73 The main effect of the within-subjects f actor, Disfluency Type, was significant ( F (1, 26) = 23.009, p < 0.001, 2 = 0.469) indicating that fo r all participants, the fr equency of repetitions and prolongations was different from the frequency of other types of disfluencies. The Group Disfluency Type interaction was also significant ( F (1, 26) = 24.893, p < 0.001 2 = 0.489) indicating that that the frequenc y of repetitions and prolongati ons versus other types of disfluencies for one study group was different from that freque ncy for the other study group. The main effect of the between-subjects factor Group was also significant ( F (1, 26) = 38.613, p < 0.001 2 = 0.598) indicating that overall, the PWS produced significantly more speech disfluencies per response ( M = 0.89) than the PWNS ( M = 0.13). To further examine the Group Disfluency T ype interaction, post-hoc paired-samples t tests were run comparing the within-subjects factor (i.e., frequency of repetitions and prolongations and other types of disfluencies) in the following e xperimental condition pairs to determine which were significantly different: Repetitions and prolonga tions versus other disf luencies produced by PWNS Repetitions and prolonga tions versus other disf luencies produced by PWS Repetitions and prolongations produced by PWNS versus r epetitions and prolongations produced by PWS Other disfluencies produced by PWNS ve rsus other disfluencies produced by PWS A Bonferroni-corrected alpha was used to cont rol Type I error (alpha levels .05/4 t tests = .0125). The results indicated that the PWS produ ced significantly more repetitions and prolongations ( M = 0.13) than other disfluencies ( M = 0.29) per response following the picture types used in the present study, ( t (13) = 5.860, two-tailed p < 0.001). The comparison for the PWNS was not significant indicating that the PWNS produced a comparable number of repetitions and prolongations a nd other types of disfluencies in responses following the

PAGE 74

74 picture types used in the presen t study. In addition, the results indicated that when comparing the two groups in terms of the types of disfluenci es they produced, the PW NS produced significantly fewer repetitions and prolongations ( M = 0.12) per response than the PWS ( M = 1.47) ( t (13) = -7.139, two-tailed p < 0.001), but the difference between th e groups in the numbers of other disfluencies was not stat istically significant (PWNS M = 0.14, PWS M = 0.31). The main effect for the within-subjects f actor, Picture Type, approached significance ( F (2, 52) = 3.007, p = 0.058) and the Picture Type Group interaction ( F (2, 52) = 1.295, p = 0.283) was not significant, indicating that although the PWS produced more speech disfluencies than PWNS, all participants seemed to produce overa ll fewer speech disfluencies per response following two-clause primes ( M = 0.45) than speech disfluenci es in responses following both transitive ( M = 0.54) and dative ( M = 0.54) primes. There was a significant Disfluency T ype Picture Type interaction ( F (2, 52) = 3.377, = 0.042, 2 = 0.115). However, the Disfluency Type Picture Type Group interaction was not significant ( F (2, 52) = 1.446, p = 0.245) indicating that although pict ure type seemed to affect fluency, it did not seem to affect the type of speech disfluency differen tly between study groups. To further examine the Disfluency Type Picture Type interac tion, post-hoc pairedsamples t tests were run comparing the within-subject s factor (i.e., frequenc y of repetitions and prolongations versus other types of disfluenci es) across the three prim e types to determine which were significantly different: Repetitions and prolongations in responses following Transitive versus Dative pictures Repetitions and prolongations in responses following Transitive versus Two-Clause pictures Repetitions and prolongations in responses following Dative versus Two-Clause pictures Other disfluencies in responses foll owing Transitive versus Dative pictures

PAGE 75

75 Other disfluencies in res ponses following Transitive versus Two-Clause pictures Other disfluencies n responses follow ing Dative versus Two-Clause pictures A Bonferroni-corrected alpha was used to control Type I error (alpha levels .05/6 t tests = .008). The result of only one paired comparison was si gnificant, indicting that participants in the two study groups produced significantly fewer speech disfluencies in respons es to the Transitive picture types ( M = 0.45) than in responses to Two-Clause picture types ( M = 0.54) ( t (27) = 2.945, two-tailed p = 0.007). In summary the results of the Picture Type an alyses indicated that when responding to the transitive, dative, and two-cl ause pictures examined in the present study, PWS produced significantly more speech disfluencies than PWNS. Additiona lly, participants in the two study groups produced significantly more overall disflu encies in responses following the two-clause primes than in responses following the transitive and dative primes. Priming Analyses Similar to the procedures used in the fluency analysis descri bed above, three analyses were used to answer the priming related questions. Reca ll that these analyses were based on the three experimental conditions (i.e., Tran sitive, Dative, and Two-clause). The first analysis examined the percentage with which passive responses were produced following active, locative, and passive primes in the Transitive experiment. The second analysis examined the percentage with which prepositional object dative responses were produced following double-object, objectcomplement, and prepositional object dative primes in the Dative experiment. The third analysis examined the percentage with which embe dded responses were produced following rightembedded, center-embedded and conjoined resp onses in the Two-Clause experiment. The remainder of this section presents the results of these three analyses.

PAGE 76

76 Transitive Figure 3-3a shows the percentage with wh ich passive responses were produced by participants in the two groups fo llowing the presentation of active, locative, and passive primes. As can be seen, the observed percentage of passi ve responses overall a nd passive responses to passive primes for the PWNS was somewhat higher ( M = 23%) than it was for the PWS ( M = 12%). As will be shown in this section, the results of the statistical analyses examining these overall and specific differences in the percentage of passive re sponses between the groups were not statistically significant. Priming-related analyses for the Transitive e xperiment were conducted using a 2 (Group) x 3 (Prime Type) mixed model analysis of va riance (ANOVA). Group (PWNS, PWS) was the between-subjects factor, and Prim e Type (Active, Locative, and Passive) was the within-subjects factor. The dependent variable wa s the percentage of sentences produced as full passives (i.e., the responses had to include a form of to be plus a by phrase) in the Transitive experiment. Results showed no main effect for the within-subjects factor Prime Type ( F (2, 52) = 0.845, p = 0.435), no Prime Type Group interaction ( F (2, 52) = 0.498, p = 0.611) and no main effect for the between-subjects factor Group ( F (1, 26) = 0.813, p = 0.375) indicating that participants in the two study groups produced a comparable numbe r of passive responses after active, locative, and passive primes. Dative Figure 3-3b shows the percentage with which prepositional dative responses were produced by participants in th e two groups following the presen tation of a double-object, objectcomplement, and prepositional-object dative primes. As can be seen, the observed percentage of prepositional-object responses to other dative responses for the PWS was somewhat higher for two of the dative prime conditions ( Preposit ional-object responses following double-object

PAGE 77

77 primes M = 43% and prepositional-object responses following prepositional-object primes M = 54%) than they were for the PWNS ( M = 18% and 49% respectively). and prepositional-object responses following prepositional-object primes M = 54%). As will be shown in this section, the results of a number of the statistical analyses examining these overall and specific differences in the percentage of prepositional-object responses between the groups were st atistically significant Priming-related analyses for the Dative experi ment were conducted using a 2 (Group) x 3 (Prime Type) mixed ANOVA. Group (PWNS, PWS) was the between-subjects factor, and Prime Type (Double-Object, Object-Complement, Prepos itional-Object dative) was the within-subjects factor. The dependent variable wa s the percentage of prepositiona l object dative responses in the Dative experiment. Results showed a significant main effect for the within-subjects factor, Prime Type, ( F (2, 52) = 9.455, p < 0.001, 2 = 0.267), and a Prime Type Group interaction ( F (2, 52) = 3.914, p = 0.026, 2 = 0.131) indicating that participants in one study group produced more prepositional-object dative respons es following at least one prime t ype than participants in the other study group. To further examine the main effect of Prime Type, post-hoc paired-samples t tests were run comparing the within-subjects factor (i.e., the percentage of prepositional-object dative responses) in the following priming type pairs to determine which were significantly different: Double-object versus object-complement primes Double-object versus prepositional-object responses Object-complement versus pr epositional-object responses A Bonferroni-corrected alpha was used to control Type I error (alpha levels = .05/3 t tests = .016). The results indicated that all particip ants produced significantly more prepositionalobject responses following pre positional-object primes ( M = 52%) than following object-

PAGE 78

78 complement primes ( M = 23%) ( t (27) = 4.191, two-tailed p < 0.001) and following doubleobject primes ( M = 33%) ( t (27) = 4.191, two-tailed p = 0.016) To further examine the Prime Type Group in teraction, post-hoc pair ed-samples t tests were run comparing the within-subjects factor (i.e., the percentage of prepositional object dative responses) in the following priming type pairs to determine which were significantly different: Responses produced by PWNS following double-obj ect versus object-complement primes Responses produced by PWNS following double-objec t versus prepositional-object primes Responses produced by PWNS following object-complement versus prepositional-object primes Responses produced by PWS following double-ob ject versus object-complement primes Responses produced by PWS following double-obj ect versus preposi tional-object primes Responses produced by PWS following objectcomplement versus prepositional-object primes Responses produced by PWNS versus PWS following double-object primes Responses produced by PWNS versus PWS following object-complement primes Responses produced by PWNS versus PWS following prepositional-object primes A Bonferroni-corrected alpha was used to control Type I error (alpha levels = .05/9 t tests = .005). Pertaining to the PWNS, non e of the results of the paired comparisons were significant. In detail, although the PWNS produced relatively fewer prepositional-object responses following double-object primes (18%) than following prep ositional-object primes (49%), the difference between the two responses was not statistically significant ( t (13) = -2.440, two-tailed p = 0.030). Similarly, and although the PWNS produced fewer prepositional-object responses following object-complement primes (30%) than following prepositional-object primes, the difference between the two responses for the PWNS was not statistic ally significant ( t (13) = -2.181, twotailed p = 0.048). Thus the results indicated that although the PWNS produced more

PAGE 79

79 prepositional-object responses following prepos itional-object primes and following objectcomplement primes than following double-object pr imes, the percentage of prepositional-object dative responses produced by the PWNS was not si gnificantly affected by th e type of the dative primes preceding those responses. Alternately, the results of only one paired-comparison for the PWS were significant. Namely, PWS produced significantly fewer prepositional-object responses following object-complement primes (16%) than prepositional-object responses following prepositional-object primes (54%) ( t (13) = -3.762, two-tailed p = 0.002). The result of another paired-comparison (pre positional-object responses fo llowing double-object primes versus following object-complement primes) only approached significance ( t (13) = 3.251, twotailed p = 0.006). Thus, for the PWS, the percentage of prepositional-object responses was affected by the type of dative prime preceding those responses. Specifically, the PWS produced the significantly more preposit ional-object responses following prepositional-object primes (54%) than following object-complement primes (16%). Although the PWS seemed to produce a more prepositional-object responses following pr epositional-object than following double-object primes, the difference between the frequencies of prepositional-object responses for this examined comparison was not statistically signifi cant. In terms of comparisons between the two groups, none of the results of the paired-tests were sign ificant indicating that although there were differences between the two groups in the fre quency of prepositional-object responses following the examined dative primes, the frequency of prepositional-object responses following the different dative primes examined in the pr esent study was comparable between the two study groups. In summary, the results of the priming type in the Dative experiment indicated that overall, participants tended to produce more preposi tional-object responses following prepositional-

PAGE 80

80 object primes than any other dative primes. The results also indicated differences between the two groups in the frequency of prepositional-obj ect responses based on the structure of the prime. Specifically for the PWNS, the type of da tive prime did not seem to significantly affect the probability using the prepositional-object pr ime in the response. In addition, the PWNS appeared to produce more prepositional-object responses following prepositional-object and object-complement primes than following double-obj ect primes, although none of the differences were statistically significant. The PWS on the ot her hand tended to produce more prepositionalobject responses following double-object primes (t he frequency was not significant compared to the frequencies following the other primes), a nd significantly more such responses following prepositional-object primes than fo llowing object-complement primes. Two-Clause Figure 3-3c shows the percentage with which various types of responses were produced by participants in the two groups fo llowing presentation of a two-clau se prime sentence. As can be seen, the observed percentage of responses with embedded clauses was different between the two groups; the PWNS produced complex sentences more frequently ( M = 61%) than the PWS ( M = 42%). As will be shown in this section, the results of the statistical analyses examining these overall and specific differences in the percentage of embedde d responses between the groups were not statistically significant Priming-related analyses for the Two-Clause experiment were conducted using a 2 (Group) x 3 (Prime Type) mixed model ANOVA. Group (PWN S, PWS) was the between-subjects factor, and Prime Type (Conjoined, Right-Embedded, and Center-Embedded) was the within-subjects factor. The dependent variable was the percenta ge of complex sentence responses in the TwoClause experiment. Results showed no significan t main effect for the within-subjects factor, Prime Type ( F (2, 52) = 0.548, p = 0.581), and no Prime Type Group interaction ( F (2, 52) =

PAGE 81

81 0.788, p = 0.460). The main effect for the betwee n-subjects factor Group approached significance ( F (1, 26) = 3.896, p = 0.059), indicating a variable response pattern across participants for embedded clauses. Summary of the Priming Analyses In the Transitive experiment, the results indi cated that overall, pa rticipants produced a comparable number of passive responses fo llowing active, locative, and passive priming sentences. In the Dative experiment, the results i ndicated that there were differences between the two groups in the percentage of prepositional-object responses based on the type of the dative prime. Specifically and for the PWS, significantly more prepositional-object responses were produced following prepositional-object primes than following object-complement primes. Finally, in the Two-Clause expe riment, the results indicated th at although their was a trend toward PWNS producing more embedded response s than PWS, prime type (i.e., conjoined sentences versus sentences with embedded clauses) did not affect the type of sentence produced by either group. SRT Analyses Before describing the analyses used to ex amine SRT, it is worth mentioning that the original plan was to look at SRT within the response types associat ed with each experiment (i.e., Transitive experiment: active, lo cative, and passive response types; Dative experiment: doubleobject, object-complement, and prepositional obj ect response types; Two-Clause experiment: conjoined, right-embedded, and center-embedded re sponse types), in addition to the congruency of these responses with the priming sentence (e .g., frequency of passive responses following a passive priming sentence). Recall also that on ly fluent responses were included in the SRT analysis. For the purposes of this analysis, fluen t responses were defined as those responses for which at least the first noun phrase contained no instances of disfluency (i.e., responses that

PAGE 82

82 followed the more lenient fluency definition described in the Methods chapter). When the fluency-related criterion (i.e., including only responses for wh ich NP1 was produced fluently) was applied across the categories of response types, many of the response t ype cells, particularly for PWS, contained no data points. Consequently data were collapsed and examined together within each of the three experiments. Collaps ing the data in this manner did not allow for examination of SRT in terms of priming. Thus, th e results reported in this subsection are for SRT across the three prime types regard less of whether or not the res ponses matched the structure of the exact the prime. Specifically, in the Transi tive experiment, overall SRTs for responses that exhibited a transitive sentence structure were examined without detailed examination and comparison of SRTs for active ve rsus locative versus passive re sponses. That is, overall SRTs for the Transitive prime type were examined wit hout comparing SRTs across the specific active, locative, or passive transitive primes. Similarl y, in the Dative experiment, overall SRTs for responses that exhibited a dativ e sentence structure were examin ed regardless of the priming condition in which they occurred. That is, overal l SRTs for the Dative prime type were examined without comparing SRTs across the specific double -object, object-complement, or prepositionalobject dative primes. Similarly, and in the Two-Cl ause experiment, overall SRTs for two-clause for responses were examined regardless of the prime condition in which they occurred. That is, SRTs for the Two-Clause prime type were examined without comparing SRTs across the specific conjoined, right-embedded, or center-embedded two-clause primes. Figure 3-4 shows the mean and standard devi ations for SRT for response types across the three experiments (i.e., responses that were transitive in the Tran sitive experiment, responses that were dative in the Dative expe riment and responses that were two-clause in the Two-Clause experiment). As can be seen, when the responses exhibited a transitive sent ence structure, SRTs

PAGE 83

83 for the PWS were somewhat shorter ( M = 2274.69 ms) than SRTs for PWNS ( M = 2372.97 ms). A similar difference and directi on of the difference in SRT betw een the two groups can be also seen when responses exhibited a dative sentence structure (PWS M = 2404.25 ms, PWNS M = 2604.89 ms) and a two-clause sentence structure (PWS M = 2638.41 ms, PWNS M = 2780.76 ms). As will be shown, the result s of the statistical analyses desc ribed in this section revealed that these observed differences between the tw o groups were not statis tically significant. Two questions were asked to answer the study questions pertaining to SRT in relation to speaker group and response type. The first quest ion considered whether there was a difference between PWS and PWNS in overa ll SRT. The second question as ked if response type (i.e., transitive, dative, two-clause) affected speech reaction times of PWS and PWNS similarly. To determine whether there were significant di fferences between the two groups in SRT, in addition to any possible difference between the groups in mean SRT for the different response types, a 2 (Group) 3 (Response Type) repeat ed measures ANOVA wa s used. Group (PWNS, PWS) was the between-subjects factor and Respons e Type (responses that were transitive in the Transitive experiment, responses th at were dative in the Dative experiment, and responses that were two-clause in the Two-Clau se experiment) was the within-subj ects factor. As stated above, the analyses included entirely fl uent responses in addition to on es that had a fluent first noun phrase. The puerility assumption was tested but was not met ( 2 (2) = 11.615, p = 0.003). Thus, the reported statistics are for the Huynh-Felt corr ection. There was a significant main effect for the within-subjects fact or, Response Type, ( F (1.58, 41.13) = 14.654, p < .001, 2 = .360). However, there was no significant main effect for the between-subj ects factor, Group ( F (1, 26) = 0.302, p = 0.587), and no significant Group Response Type interaction, ( F (1.58, 41.13) =

PAGE 84

84 .259, p = 0.721). Thus, for all particip ants, there was a difference in SRT in the responses they produced for at least one of the response types examined in the present study. To further examine the main effect of Response Type, post-hoc paired-samples t tests were run comparing the within-s ubjects factor (i.e., SRTs for transitive responses in the Transitive experiment, dative res ponses in the Dative experiment, and two-clause responses in the Two-Clause experiment) in the following re sponse type pairs to determine which were significantly different: Transitive versus dative Transitive versus two-clause Dative versus two-clause A Bonferroni-corrected alpha was used to control Type I error (alpha levels = .05/3 t tests = .016). The results of all paired comparisons were significant. In detail, th e results indicated that and for all participants, when producing sentence s in which at least the first noun phrase was fluent, SRTs for sentences exhibiting a tran sitive sentence structure were faster ( M = 2323 ms) than SRTs for sentences e xhibiting a dative structure ( M = 2504 ms) (t (27) = -4.029, two-tailed p < 0.001) and were also faster than SRTs for sentences exhibiting a tw o-clause structure ( M = 2709 ms) ( t (27) = -4.770, two-tailed p < 0.001). In addition, SRTs for sentences exhibiting a dative structure were faster ( M = 2504 ms) than SRTs for sent ences exhibiting a two-clause structure ( M = 2709 ms) ( t (27) = -2.587, two-tailed p = 0.015). In summary, the results of the SRT analyses indicated that for all participants, SRTs for transitive responses were faster than those for dativ e responses and SRTs for dative responses were faster than those for two-clause responses. To address the possibility that response leng th (in words) may have affected SRT across the three response types (i.e., Tr ansitive, Dative, and Two-Clause ) (See Figure 3-5 for the mean

PAGE 85

85 and standard deviation for utterance length (in word s) in the responses pro duced across the three experiments), a 2 (Group) 2 (Response Type) analysis of covariance (ANCOVA) was used. Group (PWNS, PWS) was the betweensubjects f actor, Response Type (tra nsitive responses in the Transitive experiment, dative responses in the Dative experiment, and two-clause responses in the Two-Clause experiments) was the within -subjects factor, and Response Length (in words) was the covariate. The dependent variable was m ean SRT for transitive, dative, and two-clause responses. As with the preceding analysis, this an alysis included entirely fluent responses in addition to ones that had a flue nt first noun phrase. The puerility assumption was tested but was not met ( 2 (2) = 11.425, p = 0.003). Thus, the reported st atistics are for the Huynh-Felt correction. There was no significan t main effect for the within-subj ects factor Response Type ( F (1.64, 40.99) = 2.470, p = 0.107), no Response Type Group interaction ( F (1.64, 40.99) = 0.267, p = 0.723), no Response Type Re sponse Length interaction ( F (1.64, 40.99) = 1.646, p = 0.208), and no main effect of the covariate Response Length ( F (1,260) = 0.067, p = 0.798). Thus, for all participants, ther e were no differences in SRTs among the examined response types when differences in response length across experiments were considered.

PAGE 86

86 Figure 3-1 Mean number and standard error for repetitions and prolonga tions (Rep.& Pro.) and other types of disfluencies (Other) in responses produced by persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when the priming sentences were (a) Transitive (Active, Loca tive, Passive), (b) Dative (Double-Object = D.obj., Object-complement = Obj.-comp., Prepositional-object = Prep.-obj.), and (c) Two-Clause (Conjoined, Right-embe dded = R-emb., Center-embedded = C.emb.).

PAGE 87

87 (a) Transitive Prime Effect on Fluency 0 0.4 0.8 1.2 1.6 2 Rep.& Pro. OtherRep.& Pro. OtherRep.& Pro. Other ActivePassiveLocative Transitive Prime TypeFrequency of Disfluenci e Per Response PWNS PWS (b) Dative Prime Effect on Fluency 0 0.4 0.8 1.2 1.6 2 Rep.& Pro. OtherRep.& Pro. OtherRep.& Pro. Other D.-obj.Obj.-comp.Prep.-obj. Dative Prime TypeFrequency of Disfluenci e Per Response PWNS PWS (c) Two-Clause Prime Effect on Fluency 0 0.4 0.8 1.2 1.6 2 Rep.& Pro. OtherRep.& Pro. OtherRep.& Pro. Other ConjoinedR-embed.C-embed. Two-Clause Prime TypeFrequency of Disfluenci e Per Response PWNS PWS

PAGE 88

88 Effect of Picture Type on Fluency 0 0.4 0.8 1.2 1.6 2 Rep.& Pro. OtherRep.& Pro. OtherRep.& Pro. Other TransitiveDativeTwo-Clause Picture TypeFrequency of Disfluenci e Per Response PWNS PWS Figure 3-2. Mean number and standard error fo r repetitions and prolong ations (Rep.& Pro.) and other types of disflu encies (Other) in response s produced by persons who do not stutter (PWNS; N = 14) and persons w ho stutter (PWS; N = 14) when the priming pictures were (a) Transitive (b) Dative, and (c) Two-Clause.

PAGE 89

89 Figure 3-3. Mean percentage and standard erro r for responses produced by persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when the priming sentences were (a) Transitive (Active, Loca tive, Passive), (b) Dative (Double-Object, Object-complement, Prepositional-object), and (c) Two-Clause (Conjoined, Rightembedded, Center-embedded).

PAGE 90

90 (a) Effect of Transitive Prime Type on Response 0 10 20 30 ActivePassiveLocative Prime TypePercentage of Passiv e Responses PWNS PWS (b) Effect of Dative Prime Type on Response 0 10 20 30 40 50 60 70 Double-object.Objectcomplement Prepositionalobject Prime TypePercentage of Prepositional object Responses PWNS PWS (c) Effect of Two-Clause Prime Type on Response 0 20 40 60 80 100 ConjoinedRight-embeddedCenter-embedded Prime TypePercentage of Embedde d Responses PWNS PWS

PAGE 91

91 Effect of Response Type on SRT 2000 2500 3000 3500 TransitiveDativeTwo-Clause Response TypeMean SRT PWNS PWS Figure 3-4 Mean and standard error for speech reaction time (SRT) for responses produced by persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) when those responses were (a) Transitive in the Transitive experi ment, (b) Dative in the Dative experiment, and (c) Two-Clau se in the Two-Clause experiment. Response Length in Words 5 10 15 TransitiveDativeTwo-Clause Response TypeWords Per Response PWNS PWS Figure 3-5 Mean and standard deviation for number of words produced by persons who do not stutter (PWNS; N = 14) and persons who stutter (PWS; N = 14) for (a) Transitive responses in the Transitive experiment (b) Dative responses in the Dative experiment, and (c) Two-Clause respons es in the Two-Clause experiment.

PAGE 92

92 CHAPTER 4 DISCUSSION Three main questions were asked in this study: (1) Does syntactic structure affect speech fluency in persons who stutter a nd persons who do not stutter? (2) Does syntactic structure affect speech reaction time in persons who stutter and pe rsons who do not stutter? (3) Does syntactic priming affect speech reaction time in persons who stutter and persons who do not stutter? and (4) Does syntactic priming affect speech fluenc y in persons who stutter and persons who do not stutter? These questions were motivated by theori es which suggest a relationship between stuttering and difficulties in the retrieval and/or formulation of syntactic information during speech production. This study was designed to test these theories and to extend prior research by examining the relationship between syntactic stru cture, speech fluency, and speech reaction time in persons who stutter. Fourteen adults who stutter and fourteen adults who do not stutter participated in the study. Responses were elic ited using a syntactic priming paradigm during which participants repeated sent ences and then described pictures depicting actions that can be described using transitive, dative, and two-clause sentence fo rms. In this chapter, I remind the reader of the main study findings and discuss the general implicati ons for those findings in light of previous research. Accordingly, the reminder of th is chapter is designed as follows. In the first section, the results of the fluenc y analyses and the fluency in re lation to priming are presented and discussed. In the second sect ion, the results of the primi ng analyses are presented and discussed. In the third section, the results of the SRT analyses and the SRT in relation to response type are presented and di scussed. In the fourth and fina l section, the limitations of the present study in addition to suggestions for future research are presented.

PAGE 93

93 The Effect of Syntactic Structure on Fluency Recall that in the present study, the purpose of using the above presented syntactic priming paradigm was to elicit responses reflecting a gr adual increase in syntactic complexity. Namely, as defined in the present study, tr ansitive sentences (i.e., active, lo cative, passive) reflected a low level of syntactic complexity, dative senten ces (i.e., double-object, object-complement, prepositional-object) reflected an intermediate level of syntac tic complexity, and two-clause sentences (i.e., conjoined, right-embedded, center-e mbedded) reflected a hi gh level of syntactic complexity. The main finding from the fluenc y analyses indicated that although the PWS produced significantly more speech disfluencies than the PWNS, the syntactic structure (i.e., syntactic complexity) of the priming sentences us ed in the present study di d not seem to affect fluency differently between the two study groups. Th is result is consiste nt with findings from previous studies that have examined the effect of syntactic complexity upon the speech fluency of adolescents and adults w ho stutter (e.g., Silverman and Ratner, 1997; Logan, 2001). This result, however, is not consistent with a numbe r of previous findings from studies involving young children and adults who stutter. In those stud ies, results have pointed toward a significant relationship between the syntactic complexity of a sentence and the fluency with which it is spoken (e.g., Bernstein Ratner, 1997; Bosshardt, Ballmer, & De Nil, 2002; Logan and Conture, 1995). It is interesting to consider the underlying factors for the difference in the effect of syntactic complexity on speech production betw een earlier reports indicating an effect in children who stutter and the result s of the present study. One explana tion pertains to the fact that almost all the research done with children who st utter to explore the eff ect of varying syntactic structure on speech production has examined this relationship in preschool and young school age children. As suggested in a number of previous studies (e.g., Silverman and Ratner, 1997; Logan,

PAGE 94

94 2001, Yaruss, 1999), the syntactic structures used in experiments with younger speakers might be developmentally challenging to those childre n. For instance, complex syntax forms may take young children a relatively long time to retrieve or assemble and th is in turn could place added stress on the speech motor system. An alternate possibility is that the children who stutter are prone to making excessive syntactic formulati on during the sentence formulation process. Any such errors would seemingly have to occur pr ior to overt speech production, however, because children who stutter have not been found to produ ce substantially more synt actic errors in their spoken language than children who do not stutter. In any case, it is logical to assume that with age and increased competency in ones native language, what might have been considered a challenging linguistic form for a younger age group might not pose much challenge in adolescence or adulthood. Along the same lines and within the framewor k of earlier presented models of stuttering (e.g., the Demands and Cap acities Model), increased syntactic complexity could be more demanding on childrens speech a nd language systems (as compared to adults) and thus, increased percentage of speech di sfluency in responses produced by children who stutter could be viewed as resulting from inab ility of their language (and probably speech) systems to handle such demands during ongoing speech. As Yaruss (1999) suggested, longer and/or more syntactically complex sentences c ontain more information units than shorter and less syntactically complex sentences. When proce ssing the more syntactically complex sentence structures, more such information units may need to be held in short-term memory until decisions about the target utterance are made. As Yaruss explained, this in turn might exert more demands on the processing system resulting in a hi gher percentage of disfluencies in the more syntactically complex responses than in the less syntactically comp lex ones. Thus, it is

PAGE 95

95 reasonable to suggest that from a developmental poi nt of view, it is not surprising that syntactic structures that have been repor ted as challenging to children ar e not as challenging to adults. Although syntactactic complexity does not appear to affect overt fluency in AWS, it does seem to affect other types of functions. For exam ple, Bosshardt et al. (2002) reported that when speakers in their study (i.e., adul ts who stutter and adults who do not stutter) were asked to generate and produce sentences during a singletask (using two given nouns to generate a sentence) versus dual-task paradigm (using two given nouns to generate a sentence and at the same time decide if the two nouns are in the sa me noun category or if they rhyme) both adults who do and do not stutter exhibited a reduced pe rcentage of accurate rhyming and category decisions. In addition, persons who stutter pr oduced significantly fewer propositions (i.e., a predicate and its argument(s)) unde r the dual-task than persons who do not stutter. The results were taken to indicate a vulne rability in the speechproducti on systems of persons who stutter which is observed during attention-demanding ta sks such as the secondary rhyme and category decision tasks used in that study. Similarly, Cuadrado et al. (2003) reported both neuralprocessing and response accuracy differences be tween persons who stutter and persons who do not stutter during online grammati cality judgment tasks. Specifically, and when asked to decide online (i.e., decide during or immediately fo llowing a response) whether the verbs in a syntactically simple versus a syntactically complex senten ce indicted a grammaticality agreement, the percentage of accurate gramma ticality judgments for persons who stutter was significantly lower than for participants who do not stutter for grammaticality violations that occurred in the more syntactically complex sentences. The results of the grammaticality judgment task in addition to resu lts of ERP measures collected during those tasks were taken by

PAGE 96

96 Cuadrado et al. (2003) to indica te that neural mechanisms reflecting postlexical analysis, specifically for syntactic reanalysis, operate differently in (PWS)) (p.974). It is worth mentioning that the study tasks in Bosshardt et al. (2002) involved generation and production of the responses while in Cuad rado et al. (2003) no speech production was involved required. Thus under task s that are considered more demanding and even in the absence of speech production, persons who st utter seem to exhibit difficulties in sentence planning and production than persons who do not stutter. Thus, in summary, it se ems that results of studies examining overt speech fluency (e.g., frequency of disfluencies per response in relation to certain syntactic forms) have shown a comparable pe rformance between the PWS and the PWNS on the syntactic tasks used in those studies. On the other hand, studies of more subtle aspects of linguistic performance (e.g., those studies exam ining syntactic processing during sentence generation through ERP examination) have shown an difference between persons who stutter and persons who do not stutter. The syntactic priming paradigm used in the present study could be viewed as more challenging than mere repetition of given utterances because of the fact that it necessitates that a participant generates a novel sentence instead of repeating the sentence he or she just heard (as opposite to the data elicitation paradigm used in Logan (2003) study viewed above). In addition, the paradigm used in the present study employe d a recognition memory task which required the participants to indicate whether or not they heard or saw the target sentences before within the framework of the experiment and which served as a secondary or concurre nt task. Based on these two characteristics of the priming paradigm use d, and although no significant effect of syntactic structure was observed in the pres ent study, I believe that the syntactic priming task used in the present study elicited sentences was in some ways closer to what a speaker might produce in

PAGE 97

97 everyday speaking contexts in th at they required that a speake r to creatively produce a sentence rather than merely produce a sentence th ey had just heard or read, as was the case in studies by Silverman and Ratner (1997) and Logan (2001). Accordingly, the results of the present study could be viewed as an addition to the increasin g body of literature that argues against a role for syntactic complexity independ ent of whatever effect that syntax may have upon utterance length in the fre quency with which adults who stut ter produce overt disfluency. Two other findings for the fluency analyses reported in the present study pertain to the frequency and type of speech disfluencies produ ced (i.e., repetitions and prolongations versus other types of disfluencies). The first finding indicated that overall, the PWS produced significantly more repet itions and prolongations per response than the PWNS. This finding is not surprising based on the fact that (a) repetit ions and prolongations ar e considered central or core behaviors in stuttering (see Bloodstein, 1995 for a detailed description), and (b) group selection criterion were base d on, among other considerations the frequency with which participants produced these t ypes of disfluency in conn ected speech. The second finding indicated that partic ipants in the two study groups produced a comparable number of other types of disfluencies per response across the exam ined sentence types. This finding is consistent with results from other studies that have compared the frequency of this disfluency class between speaker who do and do not stutter. The overlap in frequency with which speakers from both groups produced other disfluencies such as interjections and revisions in c ould reflect general similarities between the two groups in (a) sp eech processing mechanisms, or (b) general language formulation aptitude (e.g., lexical selection, syntactic form ulation). Earlier studies have shown that such disfluencies are common in t ypical speakers (see Bailey and Ferreira, 2003 for a review) and, leaving aside instances when speaker s who stutter use interjections and revision as a

PAGE 98

98 strategy for concealing repetiti ons and prolongations from listeners, seem to be peripheral to the fundamental impairment that ch aracterizes development stuttering. Another nonsignificant finding rela ted to the fluency analyses warrants some discussion. In the present study, PWNS produced a comparable number repetitions and prolongations and other disfluencies in most of the experiment al conditions. This finding was somewhat unexpected because previous research on conversat ional fluency in PWNS has found that they produce roughly twice as many other disfluencies as they do repetitions and prolongations (Bloodstein, 1995). Thus, this disfluency pattern for the PWNS suggests that sentence production tasks used in the present were so mewhat challenging for the PWNS. The Effect of Picture Type on Fluency As you may recall, all partic ipants in the present study produced significantly more disfluencies in responses follo wing the two-clause primes than in responses following the transitive primes. This finding is consistent w ith the results of Silverman and Ratners (1997) study which reported an increase in the number of normal disfluencies (e .g., interjections and revisions) by adolescents who stutter and adol escents who do not stut ter (ages 10-18 years) (although in the present study, the re ported significant increase was overall, i.e., for all types of disfluencies regardless of whethe r they were repeti tions and prolongations or other types of disfluencies). In the Silverma n et al. (1997) study, the signific ant increase in o ther types of disfluencies was observed in center-embedded res ponses (or structures defined in that study as being hardest in terms of syntactic complexity) when compared to questions (defined as being easiest in terms of syntactic complexity) a nd when compared to right-embedded responses (defined as being moderate in terms of syntactic complexity). The authors also reported the frequency of repetitions and prol ongations (or as defined in that study stuttered disfluencies) was not significantly affected by syntactic comple xity. In the present study, and as described on

PAGE 99

99 the previous chapter, the frequency of both re petitions and prolongations and other types of disfluencies seemed to significan tly increase under conditions of in creases syntactic complexity. As explained in the Methods chapter, the tran sitive sentences and pi ctures used in the present study each include two arguments (an ag ent and a patient). The dative sentences and pictures on the other hand include three argumen ts (an agent, a direct patient, and a second indirect patient). The two-clause sentences and pictures include four arguments (an agent and a patient in the main clause, and an agent and a patient in the embedded clause or in the second independent clause in the case of the conjoine d primes). It can be argued that the more arguments a response includes, the more complex th at response is in term s of the computational resources involved in generating such a response under time constraints similar to the ones used in the present study. In terms of differences in fr equency of disfluencies between the two-clause and transitive responses examined in the presen t study, it is reasonable to suggest that the twoclause responses impose greater memory loads than that imposed by the ge neration of transitive and dative sentence forms. These memory loads might result from the fact that when generating right-embedded responses as in The woman pulled the man who pulled the dog, one agent has to be held in memory and when generati ng center-embedded responses as in The woman who pulled the man pulled the dog two agents have to be held in memory to be matched with their verbs. The Effect of Syntactic Struct ure on Speech Reaction Time Recall than in the only study that examined SRT in adults who stutter in relation to syntactic complexity (i.e., L ogan, 2003), the author reported th at SRT was significantly longer for the PWS than the nonstuttering controls for th ree of the four sentence types used in that study. The first finding in the present study was th at the overall SRT va lues for PWS were on average 70.42 ms shorter than the overall SRT for PWNS (mean SRT for PWNS = 2667.20 ms

PAGE 100

100 and for PWS = 2596.78 ms). This result is inconsiste nt with earlier findi ngs which reported that the overall mean SRTs for the PWS were longe r (by approximately 132 ms) than SRTs for PWNS (Logan, 2003) for the lingui stic stimuli used in that study. This finding is also inconsistent with previous findi ngs indicating that the PWS tend to take more time than PWNS to initiate vocal responses to non linguistic stimuli (see review in Logan, 2003). It is not possible to definitely determine why the present findings ar e inconsistent with thos e previous findings. It is possible that although the pa rticipant selection procedures used in the present study were designed to ensure random selection, the group that actually particip ated might represent a subgroup of persons who stutter whose speech pr oduction processes are not as affected by linguistic and motoric planning dema nds as reported for participan ts in the earlier referenced studies. A number of earlier investigations have implied that speak ers who stutter are a heterogeneous group and that generalizations about stuttering behaviors might not always valid for all persons who stutter (Bloodstein, 1995). Fo r example, Logan (2003) reported that not all participants who stutter in his study exhibited longer SRTs compar ed to participants who did not stutter. Thus, it reasonable to suggest that perh aps including many more participants in similar future investigations might ensure a better re presentation of the population of speakers who stutter and subsequently provide more details about the effect of both time constraints and syntactic complexity on speech r eaction time in those speakers. The second finding of the SRT analyses in the present study indicated that for all participants and when examining responses in which the first noun phrase was fluent, SRTs for transitive response were significantly shorter than SRTs for dative responses which in turn were significantly shorter than those for two-clause responses. However, when response length was

PAGE 101

101 controlled (i.e., covaried ) across experiments, the difference in SRT between the response types was no longer observed. Although the finding pertaining to no differe nces between the two groups in speech reaction time seems inconsistent with that re ported in Logan (2003), the difference between the reported results of the present study and Logan ( 2003) study could be attri buted to a number of factors. First, in his preparat ion of the study material, Logan el ected to increase the syntactic complexity of an utterance by generating sentences that differed in the ela boration of the subject noun phrase prior the main verb in a given senten ce. For example, two of the sentence forms used by Logan (2003) were of (a) a Determiner + Adjective + Adjective + Noun as in The long and shiny car belongs to the girl and (b) a Determiner + Noun + Re lative Clause as in The car belongs to the girl who plays soccer. The st ructure of sentence form (a) employs pre-noun modification (i.e., Determiner + Adjective + Adjec tive), while the structure of sentence form (b) employs both pre and post noun modification (i.e ., Determiner-pre, Relative Clause-post). As mentioned above, Logan reported a significant diffe rence between the two study groups in mean SRT for three of the four sentence forms used in the study (among which are SRTs for responses reflecting the structures of sent ence forms (a) and (b) above). In the present study, the increase in syntactic complexity was define d in terms of (a) the number of arguments described within a given utterance, and (b) the numbe r of independent and dependent cl auses within that utterance. In terms of the number of arguments within each sentence type, the transitive sentences included two arguments, dative sentences included three arguments, and two-clause sentences included four arguments. In terms of the number of i ndependent and dependent clauses within each sentence type, both transitive and dative sentences included only one independent clause and no dependent clause, while the two-clause senten ces included one independe nt and one dependent

PAGE 102

102 clause. As explained in the Introduction and Methods chapters of th e present study, earlier studies have suggested that the more arguments a sentence includes the more challenging that sentence is (see Shapiro, 1997 for an overview). Th e results of the SRT an alyses in the present study indicating an increase in SRT as the syntactic (and thematic or semantic) complexity of the response increased seem to support these suggestions. Although, as mentioned above, increased elabora tions of the noun phrases (and in turn of response length) within an uttera nce have been shown in earlier studies to affect speech reaction times in both typical speakers and speakers who stutter (Ferreira, 1991; Logan, 2003) such elaborations were not factored into sentence pr eparation in the present study. Recall that when response length was controlled in the analyses, the difference in SRT between the response types was no longer observed in the present study. Base d on this finding, it could be argued that although response length has been shown in previous studies to affect SRT, the effect of syntactic structure cannot be ruled out as a cont ributing factor to differences in SRTs between response types. It can be sugge sted that future research employing the syntactic priming methodology could adopt alternative ways in de fining the syntactic complexity such as combining the definition of complexity used in the present study (i.e., th e number of elements within a given sentence in addition to number of independent and dependent clauses) and that used by Ferreira (1991) and Logan (2003) (i.e., elaboration of noun-phrase elements within the target utterance, thus increasi ng response length). Adopting this definition within the framework of a syntactic priming paradigm might assist in providing additional views of any interactions between increased syntactic complexity (and in creased utterance length) on speech timing in both speakers who do not stutte r and speakers who stutter.

PAGE 103

103 Two methodological points need to be consider ed when talking about the SRT analyses reported in the present study. First, although I argu e above that the syntactic priming paradigm used in the present study involved generation of novel sentences by the participants, one could not rule out the fact that the res ponses are not entirely novel. This is based on the fact that studies employing this paradigm have consistently report ed a tendency for participants to use many of the structures in the priming sentences when they produce a response. Thus, for all responses produced in the present study, a model was always available to reuse. This in turn might have reduced the time needed by all part icipants to generate the uttera nces and thus eliminated any group differences that might have been observed had the structures in the target sentences been truly novel. One way to overcome this could be by introducing a no prime condition to the priming study material. For example, instead of all the responses being preceded by a priming sentence, some of target responses could be pr eceded by a no-prime condition. In this case, participants would have to ge nerate, on their own and without the convenience of the modeled structure in the priming sentence, a novel sentence structure and use it to describe that picture. In this case, one could argue that differences betw een adults who stutter and adults who do not stutter in initiating their sentences might be mo re pronounced. (Of course, this approach contains one major drawback the inability to contro l for response type and response length.). Nonetheless, the argument builds on the hypothesis that when asked to perform a reasonably demanding task (such as one involving the gene ration and production of a response under time pressure), subtle differences between the la nguage and speech processing systems in persons who stutter versus persons w ho do not stutter are observable when performing such a task. Indeed, studies that have reported a difference in SRT between persons who stutter and persons who do not stutter have shown no significant difference between the groups in SRT when a

PAGE 104

104 prime was provided (similar to the report in the present study); however, such differences were significant in the absence of a prim e (e.g., Anderson and Conture, 2003). Concerning the second methodological point (i.e., Data elicitation procedures and instructions) recall that in the present study, and prior to th e presentation of the study material (i.e., priming pictures), participan ts were instructed to think of and say a sentence describing the given pictures as soon as it appears on the comput er screen. This instruction was given before any material presentation and was only presented once to each participant. Recall also that on average, the study task took about 45-60 minutes to finish. The part icipants were not reminded of that instruction anytime during the time course of material presen tation. This was done to be true to the methodological procedures used by Bock a nd colleagues in their priming studies and thus be able to compare the results of the present study to the results of those priming studies. On the other hand, in the previous studies that have reported an effect of syntactic structure on SRT (i.e., Logan, 2003), participants were spec ifically instructed and freque ntly reminded to initiate their responses as fast as they could upon the presenta tion of an auditory cue (i.e., sound signal). It could be argued that the instru ctions in Logans study might be more likely to elicit between groups differences in SRT than those used in the present study and that c ould be due to the fact that time constraint was not give n a central role in that participants were not instructed to respond as fast as they can and thus they might have taken their time to initiate their responses. This in turn may have eliminated any group diffe rences and thus result ed in not observing any differences in SRT between groups in the pres ent study. One suggestion for future research examining SRT using the priming methodology with persons who stutter is to specifically instruct the participants to initia te speech as fast as they can, and perhaps, even add reminders to do that at different times during the time course of material presentation. This in turn might add

PAGE 105

105 to the time pressure and provide addition inform ation about how participants may behave under both syntactic priming conditions and under continuous time pressure. The Effect of Syntactic Pr iming on Response Type The priming analyses in the present study yielded several findi ngs. The first finding suggested that priming methodology in general and syntactic priming in specific can be used to elicit responses reflecting a vari ety of syntactic structures, in cluding structures that are less frequently observed such as datives and passives. This result was expected and is consistent with earlier results with both typi cal speakers (e.g., Bock, 1986; Bo ck & Loebell, 1990; Smith & Wheeldon, 2001) and with speakers who stutter (A nderson & Conture, 2004; Burger & Wijnen, 1999; Hartfield & Conture, 2006; Melnick, Contur e, & Ohde, 2003; Pellowski & Conture, 2005; Wijnen & Boers, 1994). As the reader may recall, the study material used in the present study was adapted from material that was developed and has been used by Bock and colleagues for almost twenty years. The senten ce structures used in the pres ent study were also ones that previous syntactic priming studies have consisten tly reported to exhibit a priming effect. Thus, it was not surprising that participants in the pr esent study showed respons e patterns that were similar to those reported in the above listed studies (e.g., Bock, 1986; Bock & Leobell, 1990) to reuse the structure of the primi ng sentences in their responses. Another finding for the priming analyses report ed in the present study indicated that for two of the three structures examined in th e present study (i.e., tran sitive and two-clause structures), participants in the two study groups did not show significant differences in the priming effect. This result is not consistent with the only study that has examined the syntactic priming effect in persons who stutter. Specifi cally, Anderson and Contur e (2004) reported that children who stutter seem to e xhibit a great syntactic priming effect than children who do not stutter. The authors argued that children who stutter might be l ess skilled in morphosyntactic

PAGE 106

106 construction processes (p. 564) th an children who do not stutter. This in turn might affect the efficiency by which they express an intended me ssage because these children might have fewer computational resources available for syntactic processing. Within this view, children who stutter might benefit from syntactic priming becau se they might take advantage of previously activated forms (i.e., those of the priming senten ces) to more effectivel y generate the intended syntactic forms which in turn might reduce the ti me as observed in shorter SRTs for the primed sentences in that group. As mentioned above, the results of the present study did not show a difference in the extent of priming effect in the two groups. The differences between the results reported in the present study a nd those reported by Anderson and Conture (2004) might be attributed to one main reason. Na mely, studies examining syntactic processing abilities in adults who stutter have consistently s hown that although syntax might pl ay a role in stuttering during the childhood years (e.g., Bernstein Ratner and Sih, 1987; Gaines, Runyan, & Meyers, 1991; Gordon, Luper, & Peterson, 1981; Logan & Conture, 1995; 1997; Wall, Starkweather, & Cairns, 1981; Yaruss, 1999) it might not be the main contri buting factor to stutte ring during adolescence and adulthood (e.g., Silverman and Ratner, 1997; Logan, 2001; 2003). The authors of the later group of studies suggested that although some synt actic structures continue to be challenging for both adolescents and adults who do not stutter, th e effect of syntax in general and syntactic complexity in specific seems to decrease as spea kers grow older and thei r linguistic competency improves. Within this view, it could be argued that children who stutter may exhibit subtle speech production deficits and the presence of a prime in the context preceding the production of given structures may provide a priming boost and thus assist in overcoming either the retrieval or encoding difficulties that might underlie spe ech production in those children. Adults who stutter on the other hand have a hi gher linguistic competency than ch ildren who stutter as a result

PAGE 107

107 of language maturation process, and thus prim ing (syntactic priming in this case) might not play as prominent a role during speech processi ng as it would in children resulting in a smaller gain of syntactic priming similar to the reported in the present study. In terms of the effect of specific syntactic pr imes on frequency of syntactic structures in responses, the results of the pres ent study indicated that the struct ure of some syntactic primes did affect the probability of the participant reusing the stru cture in those primes. The following section presents the findings for the transitive, da tive, and two-clause prim ing structures used in the present study and discusses these resu lts in comparison with earlier findings. The first finding from the transi tive priming experiment indicat ed that for all participants, the structure of the transitive priming sentence di d not seem to significantl y affect the frequency with which participants produced passive responses. Specifically, fo r all participants 18% of the responses following passive primes were passive 18% following active prim es were passive, and 13% following locative primes were passive. In addition, and as can be observed from these percentages, the frequency of passive responses following locative primes was lower than that following both passive and active primes. These findings seem inconsistent with the ones reported by Altmann et al. (2006) and Bock et al. (1990) who re ported an overall significant increase in passive responses after passive primes in both older a nd younger speakers. The findings also seem inconsistent with those report ed by Bock et al. (1990) who suggested that the exposure to locative primes significantly increases the probability of reus ing a passive prime in the response. As the reader may recall, half the participants in the present study were college age (7 out of 14 participants). Altmann et al. (2006) ha d explained that participants in Bock et als study might have produced an overall high pr oportion of passive re sponses and passive responses following both passive and locative prim es because those participants were college

PAGE 108

108 age, and may have experienced frequent exposure to passive structures in lectures, thus making those structures more familiar to those participants and resulting in a higher probability of using those passive structures in their responses. Sti ll, in the present study, the exposure to passive and locative primes did not seem to significantly increase the probability of reusing a passive structure in the response for the participants ove rall. It could be ar gued that the number of participants in the Bock et al. study was much higher than that in th e present study and thus might have resulted in a more sensitive examinati on of the transitive primes in those studies. One suggestion for future research would be to incl ude many more participants and thus increase the sensitivity of the study procedures to capture an y effect the transitive primes may have on the production of passives. Different from the results of the transitive pr ime experiment, two signi ficant findings were reported in the dative prime experiment. The firs t finding indicated that overall, participants produced significantly more prepositional-obje ct responses following prepositional-object primes than following double-object or object-co mplement primes. The second result indicated that participants in the two study groups exhibited a different patter n in terms of the structure of the sentences they produced following the ex amined dative primes. Specifically, the PWS produced significantly more prepositional-object responses following double-object primes (43%) than following object-complement primes ( 16%). The PWNS on the other hand exhibited the opposite pattern producing more prepositiona l-object responses follo wing object-complement primes (30%) than following double-object primes ( 18%). The difference in the structure of the dative response for the PWNS was not statistically significant. It is intere sting to speculate what factors might underlie the difference in the fre quency of prepositional-responses between the two groups. Recall that the dative primes which the participants repeated in the present study

PAGE 109

109 exhibited three structures. The fi rst dative prime was that of a double-object dative verb as in The actress sold the stage manager her jewe lry. The second dative prime was that of a prepositional-object dative verb su ch as in The actress sold her jewelry to the stage manager. The third prime used in the dative experiment was that of an object-complement verb as in The stage manager nominated Mary best actress. The sentences associated with the first two dative verb forms (i.e., double-object and prepositional-object) share a comm on semantic representation or theme in that both dative sentences incl ude an agent (e.g., Mary), and two objects (e.g., boy and paintbrush). The double-ob ject and object-complement prim es on the other hand share the surface noun phrase structure, in that they bot h have a first noun phrase, followed by a verb, which in turn is followed by two noun phrases. Wh en participants are ex posed to one of the above described dative primes, and if priming is driven by the thematic aspects of the primes, one would expect a comparable frequency of prepositional-object responses following both prepositional-object primes and double-object primes. However, if priming is driven by the structural aspects of the primes, one would e xpect a comparable frequency of prepositionalobject responses following both prepositional-object and object-complement primes. As reported earlier in this chapter and in the Results chap ter, the PWS produced most prepositional-object responses following prepositional-object and double-object primes, thus, it could be argued that perhaps, priming of dative sentence forms in the PWS group is more driven by the semantic aspects of the response rather th an the structural aspect. On th e other hand, the PWNS in the present study produced more prepositional-obje ct responses following prepositional-object primes and a comparable number following obj ect-complement primes. Based on the above, it could be argued that perhaps, priming dative se ntence forms in the PWNS group is more driven

PAGE 110

110 by the structural aspects of the available options and in the PWS is more driven by the thematic aspects of the available options. The results of the two-clause prime experime nt were similar to those of the transitive prime experiment and provided one main findin g. Specifically, and for all participants, the structure of the two-clause priming sentence did not seem to significantly affect the frequency by which participants produced embedded responses. Specifically, and although participants in the two study groups exhibited a tendency to produce two-clause responses following two-clause primes, the presence of the two-clause prime di d not significantly increa se the probability of using the structure of that prime in the respons e. In terms of frequency of embedded responses following the examined two-clause primes, 54% of the responses following conjoined primes were embedded, 46% following center-embedded were embedded, and 51% following rightembedded primes were embedded. This finding is c onsistent with the one reported by Altmann et al. (2006) who reported that the structure of th e complex (i.e., two-clause) priming sentence did not seem significantly to a ffect the frequency by which ol der speakers produced embedded responses. Differences Between The Two Groups On the Prestudy Tasks Recall that all particip ants in the present study finished a number of language and memory tasks as part of the participan t screening procedures. The task s were digits forward, digits backward, digit ordering, and WA IS vocabulary test. The results of the statistical analyses comparing the means of the two groups on those tasks indicated significant differences between the two groups on two of those four tests. Namely, the mean scores for the PWNS were significantly higher on th e WAIS vocabulary and the digit or dering tests. Although differences on these tests had the potentia l of confounding any potential di fferences between the groups on of several of the analyses conducted in the presen t study, the responses of pa rticipants in the two

PAGE 111

111 study groups were comparable on the majority of the examined measures and thus, did not seem to be affected by the factors driving the differe nces on the tasks described above. Additionally, it appears that whatever role memo ry would play in study tasks si milar to the ones used in the present study, and based on the fact that there were not significant diffe rences between groups on those study tasks, it is reasonable to suggest that the role memory plays in those tasks might not be strong enough to significantly affect speech pr oduction. It is interesting to speculate what may have driven those differences in scores on the prestudy language and memory tasks. One suggestion for the difference in scores on thos e tests could be based on a common issue in stuttering research. Namely, many of the person s who stutter who voluntee r to participate in stuttering research come from a population re presenting a broad range of socioeconomic and educational background. The contro l group on the other hand might in many cases (one of which is the present study) represent a narrower range of the population (i.e., college age range) and thus reflect less diversity in that regard than the other group. Although most of the participants in the present study were matched on as many factors as possible, still, it c ould be argued that the PWS did represent a broader group while the PWNS represented a diverse group. Conclusion In the present study, the syntactic priming meth odology was used to examine the effect of syntactic structure on speech fluency and speech tim ing in adults who do and do not stutter. The results suggest that alth ough there were some differences be tween the study groups in processing certain sentence types (i.e., datives) overall, s yntactic processes involved in speech planning in persons who stutter might be performed in a manne r similar to those processes in persons who do not stutter. This suggestion is based on the findings that speech initiation times and speech fluency were similarly affected in participants in the two study groups by the structure of the sentences produced by those participants. The pres ent study is one of only few studies that have

PAGE 112

112 used the syntactic priming methodology to exam ine speech production processes in adults who stutter. The results support findi ngs from a number of previous studies and show that syntactic priming is a feasible methodology to study sentence processing abil ities in persons who stutter. In addition, and because participants tended to pr oduce a variety of structur es in their responses, this methodology could assist in examining the effect of infrequent syntactic structures such as passives and datives on speech production in persons who stutter.

PAGE 113

113 APPENDIX A PRESTUDY TESTS Background Survey: Control Group Participant ID ___________________ Todays Date __________________ DESCRIPTION OF SPEECH AND LANGUAGE: 1. Have you ever had or curren tly have difficulty with: 1a. Speech? Yes ____ No ____. If yes, please describe. ___________________________________________________________________________ 1b. Oral or written language? Yes ___ No ____. If yes, please describe. ___________________________________________________________________________ 1c. Fine or gross motor coordination? Ye s____ No _____. If yes, please describe. ___________________________________________________________________________ 2. Do you feel you hear normally? Yes ____ No ____. If no, please describe. ________________________________________________________________________ 3. Do you have difficulty remembering things? Yes ____ No ____, if yes, please describe. ________________________________________________________________________ 4. Do you feel that you have normal vision? Yes ____ No ____. 5. How would you describe your current health? Excellent ____ Good ____ Fair ____ Poor ____ 6. Are you currently taking any medication? Ye s ____ No ____. If yes please describe. ________________________________________________________________________ 7. Does the medication affect your communication? ________________________________________________________________________ ACADEMIC EDUCATIONAL (skip (a) if not curre ntly attending school) 1. Do you currently attend school? Yes ____ No ____ (if no skip to (b)) a. Name of school: ____________________________________________________ b. Year in school: _____________________________________________________ c. Major area of study: _________________________________________________ d. Highest degree earned: _______________________________________________

PAGE 114

114 Background Survey: Experimental Group Participant ID ___________________ Todays Date ____________________ DESCRIPTION OF SPEECH AND LANGUAGE: 1. Have you ever had or curren tly have difficulty with speech? Yes ____ No ____. If yes, please describe. ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ 2. At what age was your problem first noticed? ___________________________________ 3. Who noticed the problem? __________________________________________________ 4. How has the problem changed since that time? __________________________________ ________________________________________________________________________ ________________________________________________________________________ 5. Do other people notice your speech problem? ___________________________________ 6. What do you believe caused the problem? _____________________________________ 7. Have you previously had your speech, language, hearing skills evaluated? Yes ____ No ____. If yes, please provide details. ______________________________ ________________________________________________________________________ ________________________________________________________________________ 8. Have you even had speech or language therapy? Yes ____ No ____. If yes: Where: _________________________________________________________________ With whom: _____________________________________________________________ Dates: __________________________________________________________________ Focus: __________________________________________________________________ Results: _________________________________________________________________ 9. Are you currently applying any st rategies that you may have learned during speech or language therapy? Yes ____ No ____. If yes please describe. ________________________________________________________________________ ________________________________________________________________________

PAGE 115

115 10. Have you ever had or currently have problems with written language? Yes ____ No ____. If yes, please describe. _________________________________ _____________________________________________________________________ 11. Have you ever had problems with fi ne or gross motor coordination? Yes ____ No ____. If yes, please describe. _________________________________ _____________________________________________________________________ 12. Do you feel you hear normally? Yes ____ No ____. If no, please describe. _____________________________________________________________________ 13. Do you feel that you have normal vision? Yes ____ No ____. If no, please describe. __________________________________________________________ 14. Do you have difficulty remembering things? Yes ____ No ____, if yes, please describe. __________________________________________________________ 15. Do any of your close family members have a history of speech, language, hearing, or neurological problems? Yes ____ No ____. If yes please explain. _____________________________________________________________________ _____________________________________________________________________ 16. How would you describe your current health? Excellent ____ Good ____ Fair ____ Poor ____ 17. Are you currently taking any medication? Ye s ____ No ____. If yes please describe. _____________________________________________________________________ 18. Does the medication affect your communication? _____________________________ ACADEMIC EDUCATIONAL 19. Do you currently attend school? Yes ____ No ____ (if no kip to (b)) e. Name of school: _________________________________________________ f. Year in school: __________________________________________________ g. Major area of study: ______________________________________________ h. Highest degree earned: ____________________________________________

PAGE 116

116 Pre-study Tests Used with Participants Who Stutter Self-Rating Scale Participant ID# _______________ Todays Date _____________ Please respond to each of the following stat ements by circling the point along the sevenpoint scale that corresponds to your impression s of your speech. If a particular item does not pertain to you, please write not applicable in the margin. 1. I participate in group-discussions with friends. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 2. I participate in group-discussions with family. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 3. I participate in group-discussions during class or at work. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 4. I stutter when talking with family members. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 5. I enjoy talking with others. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 6. My muscles feel tense when I speak. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 7. I feel out of breath when I am speaking. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never

PAGE 117

117 8. Stuttering limits my academic a nd/or professional performance. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 9. I enjoy speaking before groups of people. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 10. I stutter when talking with friends. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 11. I feel embarrassed by the way I talk. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 12. I feel nervous when I am speaking to adults. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 13. Stuttering limits my ability to communi cate with teachers and/or employees. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 14. Stuttering limits my ability to communicate with friends. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 15. I repeat sounds or syllables when I am talking. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 16. If I expect to stutter on a certain word, I will substitute another word in its place. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never

PAGE 118

118 17. I become blocked on speech sounds when I am talking. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 18. I avoid speaking with others because of my stuttering. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 19. When talking with others, I ackn owledge the fact that I stutter. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 20. My academic skills are better than average. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 21. I think that, someday, I will be able to manage my stuttering. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 22. I do whatever is necessary to hi de my stuttering from others. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 23. I feel discouraged about the way I talk. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never 24. Other people react negatively to my stuttering. 1 ---------2 --------3 ---------4 ---------5 --------6 ----------7 Always Sometimes Never Comments: ____________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________

PAGE 119

119 Participant ID# ______________ Date ______________ Severity Level ______________ Stuttering Severity Task I want you to talk for 3 minutes about a movie you saw. I want you to talk for 3 minutes about a pleasant experience. Mean duration of three l ongest disfluencies: ______________________________________ Concomitants: _____________________________________________________________ Number of disfluencies in 600 syllables: _____ ; % of disfluencies in 100 syllables: ______ Suggested severity level: _____________________________________________________ Comments: ________________________________________________________________

PAGE 120

120 APPENDIX B PRIMING AND FILLER SENTENCES AND PICTURES Active/Passive/Locative LIST A Active The students were bankrupted by the new sports complex. The tourist was confused by the blinking traffic light. The 747 was alerted by the airport's control tower. The businessman was paged by the airline ticket agent. The scientist was inspired by the apple tree. The minister was cut by the broken stained glass window. Passive The cub scouts enjoyed the camp fire. The construction worker drove the bulldozer. The missing geologist underestimated the volcano. The secretary cleaned the drinking fountain. The quarterback destroyed the jukebox in the bar. The surfer watched the stormy sea. Locative The stockbroker was sitting by the client. The woman was swimming by the jellyfish. The lumberjack was eating by the giant redwood tree. The dog was barking by the fence. The councilman was strolling by the new building. The ship was docking by the pier. LIST B Active The stockbroker impressed the client. The woman caught the jellyfish. The lumberjack struck the giant redwood tree. The dog jumped the fence. The councilman opened the new building. The ship approached the pier. Passive The cub scouts were burned by the camp fire. The construction worker was hit by the bulldozer. The missing geologist was smothered by the volcano. The secretary was splashed by the drinking fountain. The quarterback was annoyed by the jukebox in the bar. The surfer was excited by the stormy sea.

PAGE 121

121 Locative The students were working by the new sports complex. The tourist was loitering by the blinking traffic light. The 747 was landing by the airport's control tower. The businessman was waiting by the airline ticket counter. The scientist was sleeping by the apple tree. The minister was praying by the broken stained glass. LIST C Active The students tried the new sports complex. The tourist misunderstood the blinking traffic light. The 747 radioed the airport's control tower. The businessman left the airline ticket counter. The scientist examined the apple tree. The minister fixed the br oken stained glass window. Passive The stockbroker was sued by the client. The woman was stung by the jellyfish. The lumberjack was struck by the giant redwood tree. The dog was protected by the fence. The councilman was impressed by the new building. The ship was damaged by the pier. Locative The cub scouts were singing by the camp fire. The construction worker was digging by the bulldozer. The missing geologist was wandering by the volcano. The secretary was typing by the drinking fountain. The quarterback was drinking by the jukebox in the bar. The surfer was running by the stormy sea. Double-Object Dative/Prepositional Dative/Object Complement LIST A Double-Object Dative The mother read her older children a story The volunteers sold the childre n some submarine sandwiches. The candidate wrote the Re publican mayor a letter. The ambassador's secretary sent the marine a watch. The freshman took the Salvation Army some clothes. The breeder showed the young family his best dogs.

PAGE 122

122 Prepositional Dative The king promised his daughter to the triumphant knight. The choir sang a new hymn to the wedding guests. The defendant told a lie to the suspicious lawyer. The diplomat took the secret documents to the President. The president told a joke to a Russian reporter. The widow rented the upstairs rooms to student tenants. Object Complement The students named the book a classic. The theater manager nominated Mary Best Actress. The housewife considered the church her second home. That couple's son made their garage an apartment. Proctor and Gamble appointed the drugstore owner chairman. The children elected the nurses' group their favorite volunteers. LIST B Double-Object Dative The king promised the triumphant knight his daughter. The choir sang the weddi ng guests a new hymn. The defendant told the suspicious lawyer a lie. The diplomat took the Presid ent the secret documents. The president told the Russian reporter a joke. The widow rented the student tenants the upstairs rooms. Prepositional Dative The students brought a book to Stella. The actress sold her jewelry to the stage manager. The housewife mailed a check to the bankrupt church. The couple rented the garage to their neighbor's son. Proctor and Gamble sent samples of the new detergent to consumers. The children sang a song to the nurses' group. Object Complement The mother appointed the older children "official babysitters." The volunteers considered submar ine sandwiches the perfect lunch. The Republican party elected the candidate mayor. The ambassador's secretary assigned the marine Captain of the Watch. The freshmen named the Salvation Army their favorite charity. The breeder made family pets his top priority LIST C Double-Object Dative The students brought Stella a book. The actress sold the stage manager her jewelry. The housewife mailed the minister's bankrupt church a check.

PAGE 123

123 The couple rented the neighbor's son their garage. Proctor and Gamble sent consumers samples of the new detergent The children sang the nurses' group a song. Prepositional Dative The mother read a story to her older children. The volunteers sold some submar ine sandwiches to the children. The candidate wrote a lette r to the Republican mayor. The ambassador's secretary se nt a watch to the marine. The freshman took some clothes to the Salvation Army. The breeder showed his be st dogs to the young family. Object Complement The king's daughter considered th e triumphant knight her fianc. The choir considered the ne w hymn their favorite song. The suspicious lawyer deemed the defendant a liar. The diplomat made delivering th e documents his primary mission. The president appointed the Russi an reporter his joke writer. The widow deemed the upstairs room a disaster area. Conjoined/Center-Emb edded/Right-Embedded LIST A Conjoined The man spilled the wine a nd he stained the carpet. The monkey looked under the bow l and then he hid his food. The girl interrupted the boy who was talking very loudly. The gambler chose the number and he won the lottery. The mother punished the boy and then she broke the vase. The girl smelled the flower and it reminded her of her grandmother. Center-Embedded The boys that teased the girl answered all the questions. The man that wore white pants had stains on his knees. The woman that scolded the boy climbed the tree. The carpenter that hit the man was riding a bike. The boy that was annoying the woman was talking on the phone. The spy that carried the briefcas e had the confidential documents. Right-Embedded The mechanic adjusted the fitting that closed the valve. The carpenter drove the car that won the race. The man kicked the dog that was chasing the cat. The student bought a calculat or that does calculus. The woman wore a locket that was her grandmother's. The waitress threw the dart that hit the bull's eye.

PAGE 124

124 LIST B Conjoined The mechanic adjusted the fitting and then he closed the valve. The driver drove the car and he won the race. The man kicked the dog and then he chased the cat. The student bought a calculator because she did calculus. The woman wore a locket and it was her grandmother's. The waitress threw the dart and she hit the bull's eye. Center-Embedded The man that spilled the wine stained the carpet. The monkey that looked under th e bowl was hiding his food. The girl that interrupted th e boy was humming very loudly. The gambler that chose the number won the lottery. The mother that punished the boy dropped the vase. The girl that smelled the flower was very pretty. Right-Embedded The boys teased the girl who answered all the questions. The man wore white pants th at had stains on the knee. The woman scolded the boy who had climbed the tree. The carpenter hit a man who was riding a bike. The boy was annoying the woman wh o was talking on the phone. The spy carried the briefcase that had the confidential documents. LIST C Conjoined The boys teased the girl because she answered all the questions. The man wore white pants and they had stains on the knee. The woman scolded the boy because he climbed the tree. The carpenter hit a man, but he rode away on a bike. The boy was annoying the woman because she was talking on the phone. The spy carried the briefcase becaus e it held the confidential documents. Center-Embedded The mechanic that adjusted th e fitting closed the valve. The driver that drove the car won the race. The man that kicked the dog was chasing the cat. The student that bought a cal culator takes calculus. The woman that wore a lo cket was a grandmother. The waitress that threw the dart hit the bull's eye. Right-Embedded The man spilled the wine that stained the carpet. The monkey looked under the bowl that was hiding his food. The girl interrupted the boy that was humming very loudly.

PAGE 125

125 The gambler chose the number that won the lottery. The mother punished the boy that dropped the vase. The girl smelled the flower that was very pretty. Filler Sentences The books were expensive The detective noticed the cu ts on the bartender's hand It was difficult to forget the phot ographs of the concentration camp The acorns fell from the top of the oak tree. The boy seems to enjoy the ice cream. Nancy wants to redecorate the family room. It was easy to hide the money. The house took fifty years to build Henry Vlll married Jane Seymour Bob Dole used to be in the Senate. The desk is in the room. The surgeon cut himself with a scalpel. The staff burned the incriminating papers. Jackhammers are noisy. The stockbroker was angry about the rising interest rates. The freezing rain made the street slippery. There is a red spot on Jupiter. There were thirteen original colonies. The stuntman threw himself out of the window. It was tough to fail the tennis class. That Billy was unhappy was apparent from his expression. The president of UF is Bernie Machen. George Bush is the president. There were several witnesses to the three car accident. The shaken victim described her assailant. The freezing rain made the street slippery. There were a lot of earthquakes last year. The anger of the crowd was hard to understand. There are thirty players in the tournament. McDonald's is the largest restaurant chain in the world. The weather was nice. That eighteen-year old fulfilled all the requirements for a bachelor's degree. The accident was inexplicable. Working on a computer can be frightening. The football fans lined up at the gate. The refrigerator hasn't been defrosted in months. The stockbroker was angry about the rising interest rates. Liberace died after a serious illness. The blouse had a button missing. It was impossible to answer all the questions.

PAGE 126

126 Shirley MacLaine played hers elf in one of her movies. Some of the trees on University Ave. are diseased A famous scientist is appear ing on the Discovery Channel. Independence Hall is on Chestnut Street in Philadelphia. Robert E. Lee surrendered in Virginia The calculator is broken. Priming Sentence Sets and Matched Target Pictures Transitive Sentences Priming sentence set: Active: The students tried the new sports complex. Passive: The students were bankrup ted by the new sports complex. Locative: The students were working by the new sports complex. Target picture: Toy startles little girl Priming sentence set: Active: The cub scouts enjoyed the camp fire. Passive: The cub scouts were burned by the camp fire. Locative: The cub scouts were singing by the camp fire. Target picture: Ambulance hits policeman Priming sentence set: Active: The construction worker drove the bulldozer. Passive: The construction work er was hit by the bulldozer. Locative: The construction work er was digging by the bulldozer. Target picture: Sailor kicking soldier Priming sentence set: Active: The missing geologist underestimated the volcano. Passive: The missing geologist wa s smothered by the volcano. Locative: The missing geologist was wandering by the volcano. Target picture: Avalanche scares skiers Priming sentence set: Active: The secretary clea ned the drinking fountain. Passive: The secretary was spla shed by the dri nking fountain. Locative: The secretary was t yping by the drinking fountain. Target picture: Train runs into bus Priming sentence set: Active: The quarterback destr oyed the jukebox in the bar. Passive: The quarterback was annoyed by the jukebox in the bar. Locative: The quarterback was drinking by the jukebox in the bar. Target picture: Girl kicking a boy

PAGE 127

127 Priming sentence set: Active: The surfer watched the stormy sea. Passive: The surfer was excited by the stormy sea. Locative: The surfer was running by the stormy sea. Target picture: Lightning strikes church Priming sentence set: Active: The stockbroker impressed the client. Passive: The stockbroker was sued by the client. Locative: The stockbroker was sitting by the client. Target picture: Dog chases mailman Priming sentence set: Active: The woman caught the jellyfish. Passive: The woman was stung by the jellyfish. Locative: The woman was swimming by the jellyfish. Target picture: Horse kicks cow Priming sentence set: Active: The tourist misunderst ood the blinking traffic light. Passive: The tourist was confused by the blinking traffic light. Locative: The tourist was loiter ing by the blinking traffic light. Target picture: Hydrant squirts fireman Priming sentence set: Active: The 747 radioed the airport's control tower. Passive: The 747 was alerted by th e airport's control tower. Locative: The 747 was landing by the airport's control tower. Target picture: Turtle squirts a mouse Priming sentence set: Active: The businessman left the airline ticket counter. Passive: The businessman was page d by the airline ticket agent. Locative: The businessman was wa iting by the airline ticket counter. Target picture: Wrecking ball destroys building Priming sentence set: Active: The minister fixed th e broken stained glass window. Passive: The minister was cut by the broken stained glass window. Locative: The minister was pray ing by the broken stained glass. Target picture: Tornado destroys barn Priming sentence set: Active: The scientist examined the apple tree. Passive: The scientist was inspired by the apple tree.

PAGE 128

128 Locative: The scientist was sleeping by the apple tree. Target picture: Woman kisses man Priming sentence set: Active: The councilman opened the new building. Passive: The councilman was impressed by the new building. Locative: The councilman was strolling by the new building. Target picture: Rock hits boy on head Priming sentence set: Active: The ship approached the pier. Passive: The ship was damaged by the pier. Locative: The ship was docking by the pier. Target picture: Boy is saving a girl from drowning Priming sentence set: Active: The lumberjack stru ck the giant redwood tree. Passive: The lumberjack was struck by the giant redwood tree. Locative: The lumberjack was eating by the giant redwood tree. Target picture: Fireman rescues baby from fire Priming sentence set: Active: The dog jumped the fence. Passive: The dog was protected by the fence. Locative: The dog was barking by the fence. Target picture: Bee stings man Dative Sentences Priming sentence set: Double-Object Dative: The candidate wrote the Republican mayor a letter. Prepositional Dative: The candidate wr ote a letter to the Republican mayor. Object Complement: The Republican Party elected the candidate mayor. Target picture: Waitress give menu to man Priming sentence set: Double-Object Dative: The mother read her older children a story. Prepositional Dative: The mother re ad a story to her older children. Object Complement: The mother appointed the older children "offi cial babysitters." Target picture: Boy giving apple to teacher Priming sentence set: Double-Object Dative: The volunteers sold the children some su bmarine sandwiches. Prepositional Dative: The volunteers sold so me submarine sandwiches to the children. Object Complement: The volunteers consider ed submarine sandwiches the perfect lunch. Target picture: Guy gives guitar to musician

PAGE 129

129 Priming sentence set: Double-Object Dative: The ambassador's secretary sent the marine a watch. Prepositional Dative: The ambassador's s ecretary sent a watch to the marine. Object Complement: The ambassador's secretar y assigned the marine Captain of the Watch. Target picture: Girl hands paintbrush to boy on ladder Priming sentence set: Double-Object Dative: The freshmen t ook the Salvation Army some clothes. Prepositional Dative: The freshmen named th e Salvation Army their favorite charity. Object Complement: The freshmen named the Salvation Army their favorite charity. Target picture: Salesman shows car to a couple Priming sentence set: Double-Object Dative: The breeder s howed the young family his best dogs. Prepositional Dative: The breeder show ed his best dogs to the young family. Object Complement: The breeder made family pets his top priority. Target picture: Boy and girl give flowers to the man Priming sentence set: Double-Object Dative: The king promised the triumphant knight his daughter. Prepositional Dative: The king promised his daughter to the triumphant knight. Object Complement: The king's daughter cons idered the triumphant knight her fianc. Target picture: Cowboy gives hat to clown Priming sentence set: Double-Object Dative: The choir sang the wedding guests a new hymn. Prepositional Dative: The choir sang a new hymn to the wedding guests. Object Complement: The choir consider ed the new hymn their favorite song. Target picture: Girl reads to boy Priming sentence set: Double-Object Dative: The defendant to ld the suspicious lawyer a lie. Prepositional Dative: The defendant to ld a lie to the suspicious lawyer. Object Complement: The suspicious la wyer deemed the defendant a liar. Target picture: Waitress served drinks to man Priming sentence set: Double-Object Dative: The diplomat t ook the President the secret documents. Prepositional Dative: The diplomat took th e secret documents to the President. Object Complement: The diplomat made de livering the documents his primary mission. Target picture: Boy and girl show pi cture to the teacher Priming sentence set: Double-Object Dative: The president told the Russian reporter a joke. Prepositional Dative: The president told a joke to a Russian reporter.

PAGE 130

130 Object Complement: The president appointed the Russian reporter his joke writer. Target picture: Lawyer shows gun to judge Priming sentence set: Double-Object Dative: The widow rented the student tenants the upstairs rooms. Prepositional Dative: The widow rented th e upstairs rooms to student tenants. ------: The widow deemed the upstairs room a disaster area. Target picture: Man is passing pitcher to woman Priming sentence set: Double-Object Dative: The students brought Stella a book. Prepositional Dative: The stude nts brought a book to Stella. Object Complement: The students named the book a classic. Target picture: Nurse gives stethoscope to doctor Priming sentence set: Double-Object Dative: The actress so ld the stage manager her jewelry. Prepositional Dative: The actress sold her jewelry to the stage manager. Object Complement: The stage manager nominated Mary Best Actress. Target picture: Boy gives valentine to girl Priming sentence set: Double-Object Dative: The housewife mailed th e minister's bankrupt church a check. Prepositional Dative: The housewife ma iled a check to the bankrupt church. Object Complement: The housewife cons idered the church her second home. Target picture: Woman throws stick to dog Priming sentence set: Double-Object Dative: The couple rent ed the neighbor's son their garage. Prepositional Dative: The couple rented the garage to their neighbor's son. Object Complement: That couple's son made their garage an apartment. Target picture: Cop gives ticket to man Priming sentence set: Double-Object Dative: Proctor and Gamble sent consumers samples of the new detergent. Prepositional Dative: Proctor and Gamble sent samples of the new detergent to consumers. Object Complement: Proctor and Gamble appointed the drugstore owner chairman. Target picture: Librarian gives book to boy Priming sentence set: Double-Object Dative: The child ren sang the nurses' group a song. Prepositional Dative: The childre n sang a song to the nurses' group. Object Complement: The children elected the nurses' group their favorite volunteers. Target picture: Waitress gives menu to man

PAGE 131

131 Conjoined-Embedded Sentences Priming sentence set: Conjoined: The man spilled the wi ne and he stained the carpet. Center-Embedded: The man that spilled the wine stained the carpet. Right-Embedded: The man spilled th e wine that stained the carpet. Target picture: Boy chasing cat, cat chasing duck Priming sentence set: Conjoined: The monkey looked under the bowl and then he hid his food. Center-Embedded: The monkey that l ooked under the bowl was hiding his food. Right-Embedded: The monkey looked unde r the bowl that was hiding his food. Target picture: Boy touching girl, girl touching monkey Priming sentence set: Conjoined: girl interrupted th e boy that was humming very loudly. Center-Embedded: The girl that interr upted the boy was humming very loudly. Right-Embedded: The girl interrupted the boy who was talking very loudly. Target picture: Cat and boy kicking girl Priming sentence set: Conjoined: The gambler chose the number and he won the lottery. Center-Embedded: The gambler that chose the number won the lottery. Right-Embedded: The gambler chose the number that won the lottery. Target picture: Dog chasing girl, girl chasing boy Priming sentence set: Conjoined: The mother punished th e boy and then she broke the vase. Center-Embedded: The mother that punished the boy dropped the vase. Right-Embedded: The mother punished the boy that dropped the vase. Target picture: Dog biting cat, cat biting girl Priming sentence set: Conjoined: The girl smelled the flower and it reminded her of her grandmother. Center-Embedded: The girl that sm elled the flower was very pretty. Right-Embedded: The girl smelled th e flower that was very pretty. Target picture: Man pulling woman, woman pulling dog Priming sentence set: Conjoined: The mechanic adjusted the fitting and then he closed the valve. Center-Embedded: The mechanic that ad justed the fitting closed the valve. Right-Embedded: The mechanic adjusted the fitting that closed the valve. Target picture: Bear leading girl girl leading boy Priming sentence set: Conjoined: The driver drove the car and he won the race.

PAGE 132

132 Center-Embedded: The driver th at drove the car won the race. Right-Embedded: The driver drov e the car that won the race. Target picture: Monkey patting girl, cat jumping on girl Priming sentence set: Conjoined: The man kicked the dog and then he chased the cat. Center-Embedded: The man that ki cked the dog was chasing the cat. Right-Embedded: The man kicked the dog that was chasing the cat. Target picture: Boy watching bear spying on bird Priming sentence set: Conjoined: The student bought a calculator because she did calculus. Center-Embedded: The student that bought a calculator takes calculus. Right-Embedded: The student bought a calculator that does calculus. Target picture: Girl pulling cat and chasing baby Priming sentence set: Conjoined: The woman wore a locket and it was her grandmother's. Center-Embedded: The woman that wo re a locket was a grandmother. Right-Embedded: The woman wore a locket that was her grandmother's. Target picture: Girl pulling baby and dog Priming sentence set: Conjoined: The waitress threw the dart and she hit the bull's eye. Center-Embedded: The waitress that threw the dart hit the bull's eye. Right-Embedded: The waitress threw th e dart that hit the bull's eye. Target picture: Bear spying on man looking at boy Priming sentence set: Conjoined: The man wore white pants and they had stains on the knee. Center-Embedded: The man that wore white pants had stains on his knees. Right-Embedded: The man wore white pa nts that had stains on the knee. Target picture: Cat kicking girl, girl kicking boy Priming sentence set: Conjoined: The boys teased the girl b ecause she answered all the questions. Center-Embedded: The boys that teased the girl answered all the questions. Right-Embedded: The boys teased the girl who answered all the questions. Target picture: Boy trying to pat cat, cat scratching dog Priming sentence set: Conjoined: The woman scolded the boy because he climbed the tree. Center-Embedded: The woman that scolded the boy climbed the tree. Right-Embedded: The woman scolded the boy who had climbed the tree. Target picture: Woman staring at man, man watching baby

PAGE 133

133 Priming sentence set: Conjoined: The carpenter hit a ma n, but he rode away on a bike. Center-Embedded: The carpenter that hit the man was riding a bike. Right-Embedded: The carpenter hi t a man who was riding a bike. Target picture: Monkey patting boy, boy patting girl Priming sentence set: Conjoined: The boy was annoying the woman because she was talking on the phone. Center-Embedded: The boy that was a nnoying the woman was talking on the phone. Right-Embedded: The boy was annoying th e woman who was talking on the phone. Target picture: Woman pulling man and dog Priming sentence set: Conjoined: The spy carried the briefcase because it held the c onfidential documents. Center-Embedded: The spy that carried th e briefcase had the c onfidential documents. Right-Embedded: The spy carried the brief case that had the confidential documents. Target picture: Girl leading bear, bear leading boy

PAGE 134

134 LIST OF REFERENCES Alvarez, S., Yimoyines, B., Key-DeLyria, S., & Altmann, L. J. P. (November, 2006). Age differences in the stimulabili ty of sentence structure: Implications for treatment. Paper presented at the annual meeting of the Am erican Speech-Language -Hearing Association, Miami, FL. Anderson, J., & Conture, E. (2004). Sentence-s tructure priming in young children who do and do not stutter. Journal of Speech, Language and Hearing Research, 47 552. Bernstein Ratner, N., & Sih, C. C. (1987). Effects of gradual increases in sentence length and complexity on childrens disfluency. Journal of Speech and Hearing Disorders, 52 278 287. Besche, C., Passerieux, C., Segui, J., Sarfati, Y., Laurent, J. P., & Hardy-Bayle, M. C. (1997). Syntactic and semantic processing in schizophr enic patients evalua ted by lexical-decision tasks. Neuropsychology, 11 (4):498-505. Bloodstein, O. (1995). A handbook on stuttering (5th ed.). San Diego, CA: Singular. Bock, K. (1986). Syntactic pers istence in language production. Cognitive Psychology, 18, 355 387. Bock, K., & Loebell, H. (1990). Framing sentences. Cognition, 35, 1. Bosshardt, H. G. (1993). Differences between stu tterers and nonstutterers short-term recall and recognition performance. Journal of Speech and Hearing Research, 36, 286. Bosshardt, H. G., Ballmer, W., & de Nil, L. F. (2002). Effects of category and rhyme decisions on sentence production. Journal of Speech, Language, and Hearing Research, 45 844657. Bosshardt, H. G., & Fransen, H. (1996). Online se ntence processing in a dults who stutter and adults who do not stutter. Journal of Speech and Hearing Research, 39, 785. Branigan, H. P., Pickering, M. J., McLean, J. F ., & Stewart, A. (2006). The role of local and global syntactic structure in language production: eviden ce from syntactic priming. Language and Cognitive Processes, 21 974-1010. Burger, R., & Wijnen, F. (1999). Phonological en coding and word stress in stuttering and nonstuttering subjects. Journal of Fluency Disorders, 24, 91. Caruso, A. J., Chozko-Zajko, W. J., Bidinger, D. A., & Sommers, R. K. (1994). Adults who stutter: Response to cognitive stress. Journal of Speech and Hearing Research, 37 738745. Conture, E. G. (2001). Stuttering: Its nature, diagnosis, and treatment Needham Heights, MA: Allyn & Bacon.

PAGE 135

135 Cuadrado, E. M., & Weber-Fox, C. (2003). Atypical syntactic processing in individuals who stutter: Evidence from event-related brai n potentials and behavioral measures. Journal of Speech, Language, and Hearing Research, 46, 960. de Roo, E., Kolk, H., & Hofstede, B. (2003). St ructural properties of syntactically reduced speech: a comparison of normal speakers and Broca's aphasics. Brain and Language, 86 (1):99-115. Faust, M., Silber, A., & Kaniel, S. (2001). Evidence from sentence priming for an atypical language organization in the brain of dyslexic males. Laterality, 6 (1):39-56. Ferreira, F. (1991). Effects of length and syntacti c complexity on initiation times for prepared utterances. Journal of Memory and Language, 30 210-233. Gaines, N. D., Runyan, C. M., & Meyers, S. C. ( 1991).A comparison of stutterers fluent versus stuttered utterances on measur es of length and complexity. Journal of Speech and Hearing Research 34 37. Gibson, E. (1998). Linguistic complexity: locality of syntactic dependencies. Cognition, 68 176. Gordon, P. A., Luper, H. L., & Peterson, H. A. ( 1986). The effects of synt actic complexity on the occurrence of disfluencies in 5 year old nonstutterers. Journal of Fluency Disorders, 11, 151. Greenbaum, S., & Quirk, R. (1996). A students grammar of the English language (10th ed.). Essex: Longman. Hartfield, K., & Conture, E. (2006) Effects of perceptual and con ceptual similarity in lexical priming of young children who st utter. Preliminary findings. Journal of Fluency Disorders, 31 303, 324. Hartsuiker, R. J., & Kolk, H. H. (1998). Syntacti c facilitation in agrammatic sentence production. Brain and Language, 62, 221. Jarvis, B. G. (2006). DirectRT (Version 2006 2.16) [Computer Software]. New York, NY: Empirisoft Corporation. Karniol, R. (1995). Stuttering, language, and co gnition: A review and model of stuttering as suprasegmental sentence plan alignment (SPA). Psycholinguistic Bulletin, 117 1, 104-124. Kolk, H., & Postma, A. (1997). Stuttering as a co vert repair phenomenon. In R. Curlee & G. Siegel (Eds.), Nature and treatment of st uttering: New directions (2nd ed., pp. 182 203). Boston: Allyn & Bacon. Levelt, W., J., M., & Kelter, S. (1982). Su rface form and memory in question answering. Cognitive Psychology, (14) 1, 78-106.

PAGE 136

136 Logan, K. J. (2001). The effect of syntactic comp lexity upon the speech fluency of adolescents and adults who stutter. Journal of Fluency Disorders 26 85. Logan, K. J. (2003). The effect of syntactic st ructure upon speech initiation times of stuttering and nonstuttering speakers. Journal of Fluency Disorders, 28 17-35. Logan, K. J., & Conture, E. G. (1995). Length, gr ammatical complexity, a nd rate differences in stuttered and fluent conv ersational utterances of children who stutter. Journal of Fluency Disorders, 20, 35. Logan, K. J., & Conture, E. G. (1997). Tempor al, grammatical, and phono logical characteristics of conversational utte rances produced by children who stutter. Journal of Speech and Hearing Research, 40, 107. Manning, W. H. (2000). Appeal of the demands and capacities model: Conclusions. Journal of Fluency Disorders, 25 4, 377-383. Melnick, K. S., Conture, E. G., & Ohde, R. N. (2003). Phonological primi ng in picture naming of young children who stutter. Journal of Speech, Language, and Hearing Research, 46, 1428. Nippold, M. A. (1990). Concomitant speech and la nguage disorders in stuttering children: A critique of the literature. Journal of Speech and Hearing Disorders, 55, 51. Pellowski, M. W., & Conture, E. G. (2005). Lexical encoding in young children who do and do not stutter. Journal of Speech, Language, and Hearing Research, 48, 278-294. Perkins, W. H., Kent, R. D., & Curlee, R. F. (1991). A theory of neuropsycholinguistic function in stuttering. Journal of Speech and Hearing Research, 34, 734. Peters, T. J., & Guitar, B. (1991). Stuttering: An integrated appr oach to its nature and treatment. Baltimore: Williams and Wilkins. Pickering, M. J., & Branigan, H. P. (1999) Syntactic priming in language production. Trends in Cognitive Sciences, 3, 136. Postma, A., & Kolk, H. (1993). The covert repair hypothesis: Prearticulatory repair processes in normal and stuttered disfluencies. Journal of Speech and Hearing Research, 36, 472. Ratner, N. B. (1997). Stuttering: A psycholinguistic perspective. In R. Curlee & G. Siegel (Eds.), Nature and treatment of st uttering: New directions (2nd ed., pp. 99 127). Boston: Allyn & Bacon. Riley, G. D. (1994). Stuttering Severity Instrume nt for Children and Adults Austin, TX: ProEd. Siegel, G. M. (2000). Demands and cap acities or demands and performance? Journal of Fluency Disorders, 25 4,321-327.

PAGE 137

137 Silverman, S., & Ratner, N. B. (1997). Syntactic complexity, fluency, and accuracy of sentence imitation in adolescents. Journal of Speech, Language, and Hearing Research, 40, 95. Smith, A., & Kelly, E. (1997). Stuttering: A dynamic, multifactorial model. In R. F. Curlee & G. M. Siegel (Eds.), Nature and treatment of st uttering: New directions (2nd ed., pp. 204 217). Needham Heights, MA: Allyn & Bacon. Smith, M., & Wheeldon, L. (2001). Syntactic prim ing in spoken sentence productionAn online study. Cognition, 78, 123. Starkweather, C. W. (1987). Fluency and stuttering. Englewood Cliffs, NJ: Prentice-Hall. Wall, M., Starkweather, C. W., & Cairns, H. S. (1981). Syntactic influences on stuttering in young child stutterers. Journal of Fluency Disorders, 6, 283. Watkins R. V., Yairi E., & Ambrose, N. G. (1999) Early childhood stuttering II I: initial status of expressive language abilities. Journal of Speech, Language, and Hearing Research, 42 (5):1125-35. Weber-Fox, C., Spencer, R., Spruill, J., & Smith, A. (2004). Phonologic processing in adults who stutter: Electrophysiological & behavioral evidence. Journal Speech Language and Hearing Research 47 1244. Wijnen, F., & Boers, I. (1994). Phonologi cal priming effect s in stutterers. Journal of Fluency Disorders, 19, 1. Yaruss, J. S. (1999). Utterance length, synt actic complexity, and childhood stuttering. Journal of Speech and Hearing Research, 42, 329. Yaruss, J. S. (2000). The role of performa nce in the demands and capacities model. Journal of Fluency Disorders, 25 4, 347-358.

PAGE 138

138 BIOGRAPHICAL SKETCH Maisa Haj-Tas obtained a BA in English langua ge and literature and an MA is speech pathology in her home country Jordan. She was a teacher of English as a second language for several years before she came to UF in 2001 to pursue a doctoral degree is speech pathology and was a teaching assistant there until the end of fa ll, 2006. She worked on several research projects with her advisor Dr. Kenneth Logan and presented several original papers at one international and several national conventions. Since early 2007 until the end of the Summer se mester of that year, Maisa was a teaching assistant at the English Language Institute at UF where she taught readin g and writing skills to intermediate level seco nd language learners. During her years at UF, Maisa was an active me mber in the International Student Speakers Bureau and presented several talks to students at schools in Gainesville about her country and culture. She was awarded the Al ec Courtelis award for outstandi ng international student in 2006 for her contribution and activitie s at UF. She also received se veral other prestigious awards among which were the Grinter Fellowship, the Di ssertation Fellowship, in addition to several travel awards from her department, co llege, and the graduate school at UF. Upon completion of her doctoral work at UF, Ma isa plans to go back to Jordan and will join the faculty at the Un iversity of Jordan. She is proud to sa y that she is the first person in her country of about five million people with a doctora l degree in stuttering. He future plans include teaching, conducting research, and a ssisting in the fluency clinic.