<%BANNER%>

The Diagnostic Utility of a Multi-Task Verbal Fluency Paradigm in Frontal and Temporal Lobe Epilepsy

Permanent Link: http://ufdc.ufl.edu/UFE0024780/00001

Material Information

Title: The Diagnostic Utility of a Multi-Task Verbal Fluency Paradigm in Frontal and Temporal Lobe Epilepsy An Analysis of Fluency Type and Qualitative Performance
Physical Description: 1 online resource (109 p.)
Language: english
Creator: Sachs, Bonnie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: epilepsy, fluency, neuropsychology
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Epilepsy is a chronic, disabling condition affecting roughly 50 out of every 100,000 Americans. The most common site for seizure onset is in either the frontal (FLE) or temporal lobes (TLE) of the brain. Identification of the onset of seizures is important, and partly determined by cognitive test scores. Elucidating a pattern of neuropsychological test performance in FLE and TLE is complicated, but allows for more accurate identification of seizure onset, which is essential in treatment planning. Presently, these cognitive patterns are poorly defined and overlapping, partly due to the lack of specificity of our cognitive tests. The purpose of this study was to evaluate the utility of multiple measures of verbal fluency in the differential diagnosis of intractable FL and TL epilepsy. Patients in the study included pre- and post-surgical refractory epilepsy patients with either temporal (N=14) or frontal lobe epilepsy (N=7) and healthy age and education matched controls (N=20). Patients completed a battery of neuropsychological tests thought to be sensitive to FL and TL functioning, including standard semantic and phonemic fluencies and novel action and name fluencies. We found significant differences between patients and controls for action and name fluency. Patient groups were not statistically different on fluency measures, though effect sizes indicated FL patients outperformed TL patients on name fluency. A qualitative analysis of fluency (clusters and switches) only differed for patient groups on name fluency as well. Only name and semantic fluency were adequate predictors of patient group membership. We found support for the notion that all fluency measures were related to overall intellectual ability and verbal/semantic factors. However, measures of semantic and name fluency were more related to semantic abilities and phonemic and action fluency were also related to measures of executive functioning. Results of the study indicate that in a mixed pre- and post-surgical epilepsy population, phonemic, action, and semantic fluency were not specific to frontal and temporal lobe functioning. Further, qualitative assessments of fluency did not offer significant information about seizure foci. Name fluency differentiated well between patient groups and appears to be a novel measure sensitive to the integrity of the left temporal lobe.
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 Bonnie Sachs.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Bauer, Russell M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024780:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024780/00001

Material Information

Title: The Diagnostic Utility of a Multi-Task Verbal Fluency Paradigm in Frontal and Temporal Lobe Epilepsy An Analysis of Fluency Type and Qualitative Performance
Physical Description: 1 online resource (109 p.)
Language: english
Creator: Sachs, Bonnie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: epilepsy, fluency, neuropsychology
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Epilepsy is a chronic, disabling condition affecting roughly 50 out of every 100,000 Americans. The most common site for seizure onset is in either the frontal (FLE) or temporal lobes (TLE) of the brain. Identification of the onset of seizures is important, and partly determined by cognitive test scores. Elucidating a pattern of neuropsychological test performance in FLE and TLE is complicated, but allows for more accurate identification of seizure onset, which is essential in treatment planning. Presently, these cognitive patterns are poorly defined and overlapping, partly due to the lack of specificity of our cognitive tests. The purpose of this study was to evaluate the utility of multiple measures of verbal fluency in the differential diagnosis of intractable FL and TL epilepsy. Patients in the study included pre- and post-surgical refractory epilepsy patients with either temporal (N=14) or frontal lobe epilepsy (N=7) and healthy age and education matched controls (N=20). Patients completed a battery of neuropsychological tests thought to be sensitive to FL and TL functioning, including standard semantic and phonemic fluencies and novel action and name fluencies. We found significant differences between patients and controls for action and name fluency. Patient groups were not statistically different on fluency measures, though effect sizes indicated FL patients outperformed TL patients on name fluency. A qualitative analysis of fluency (clusters and switches) only differed for patient groups on name fluency as well. Only name and semantic fluency were adequate predictors of patient group membership. We found support for the notion that all fluency measures were related to overall intellectual ability and verbal/semantic factors. However, measures of semantic and name fluency were more related to semantic abilities and phonemic and action fluency were also related to measures of executive functioning. Results of the study indicate that in a mixed pre- and post-surgical epilepsy population, phonemic, action, and semantic fluency were not specific to frontal and temporal lobe functioning. Further, qualitative assessments of fluency did not offer significant information about seizure foci. Name fluency differentiated well between patient groups and appears to be a novel measure sensitive to the integrity of the left temporal lobe.
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 Bonnie Sachs.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Bauer, Russell M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024780:00001


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 E20110216_AAAAIG INGEST_TIME 2011-02-17T00:32:33Z PACKAGE UFE0024780_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 20771 DFID F20110216_AABTLH ORIGIN DEPOSITOR PATH sachs_b_Page_063.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
1c2c09e3bf66958863978e18eaa1b076
SHA-1
b5596e466575b32734a27d75bc496f119abd66fd
53500 F20110216_AABTKT sachs_b_Page_046.pro
32f5a1e6440ca6ec16e2146887fbfa5e
912921309034dd59646060ac0e5ed9f3d02a8b56
5936 F20110216_AABTLI sachs_b_Page_064.pro
997d55886f30df1f9b31e19a5b22dc56
f340c7b0ed3954ee5b04c68fec85ea708d0973bb
53124 F20110216_AABTKU sachs_b_Page_047.pro
f69c8fdfe1376accf206895f6b390082
478471c423e4da371c92289cfb964fb11f0c2206
8592 F20110216_AABTLJ sachs_b_Page_065.pro
bddac5057340882c0420bd186137556d
4468fddbc50343faba3d639b9bccfc20743bb3c1
53135 F20110216_AABTKV sachs_b_Page_048.pro
6d5a0038362690aad17d2d3a5f0eaff6
153c1822f40094fb44d7cbdb3a615a02e1f9facf
3370 F20110216_AABTLK sachs_b_Page_066.pro
f1d2db7a6f6ac3aef975c0a66cc696f3
65c3e5d8e5f13b7ed1a4b3c6065ba59587bb499f
29828 F20110216_AABTKW sachs_b_Page_049.pro
6dd92a9e310247803b8cf9768a8eb57b
4e44cfa366ecf016ea3dda790a4c02eaa905b09e
51312 F20110216_AABTLL sachs_b_Page_067.pro
1926cd77423c30a65b2466bd108cead7
dec83aa96807d47c2b1fbc8c502baea466c22530
14510 F20110216_AABTKX sachs_b_Page_050.pro
aeab61e64ee231f33aeca260d8ff41f8
eceb64ce1848c4d6f2c587be1c074a440a29f4b0
56457 F20110216_AABTMA sachs_b_Page_085.pro
311b7ee91555bbf614588a1db64843e9
9f9e330cd6042b2b086e189748a1279992fa5fe1
56962 F20110216_AABTLM sachs_b_Page_068.pro
233217cb553f892d896c6482eec00504
72cebbff723329a37e6fff7dd5c538e8785d4949
47864 F20110216_AABTKY sachs_b_Page_052.pro
624f56dc01a2128c00915da7b29beb4a
60468860cddf00dc6b4c4eced25bd6ba18ddda07
57859 F20110216_AABTMB sachs_b_Page_086.pro
2552c24ff39efc97739bcaae41e80aea
b48cda8c8ec1bec7ccae26abe98fcedcd0689222
52116 F20110216_AABTLN sachs_b_Page_069.pro
9166b4ac0e45bc4d8c56dc95f306b310
8422e905b662685d3cd3f103411d37322805528b
53774 F20110216_AABTKZ sachs_b_Page_053.pro
e86dfff57b5081220af2cd7acd601689
7294356dba7cb883064b83b7d11a818785e1c21e
56797 F20110216_AABTMC sachs_b_Page_087.pro
6172d2c4d74a5a97734d6fa103172f5e
25372390f4605d4229237d1649e63b95f46bd0ba
53246 F20110216_AABTLO sachs_b_Page_070.pro
91fea1430b1eceb8670b663d9bb4ad46
8ca180caf2102a3f0c2e7ef195b7dbbb53b2a345
57160 F20110216_AABTMD sachs_b_Page_088.pro
502182aa6d5f5baa0c7cfc341bb1ee7d
2098c9e5517e6e9d3aae104cae33c2a5cf6e174f
54143 F20110216_AABTLP sachs_b_Page_071.pro
a80a2458212760e25ea4bbd23a5e0b5d
e5a90a4c734bfa4af38aed829059cde283432e03
52980 F20110216_AABTME sachs_b_Page_089.pro
d888e8d4b3f37045fb0f5ea68db638d9
7007ffaa2f2c9ea4a7b9562fa83c6d5fceadae49
53704 F20110216_AABTLQ sachs_b_Page_072.pro
c7e5d66fa5f05599fd95bb7cc0dbbd5f
afa49a8ec8e1c33042eaeb598b3358ad36bbbc27
55020 F20110216_AABTMF sachs_b_Page_090.pro
b73288ff4897a7ef44761ebf0d91fdbf
c8f61d7719fa9c7402b5831630e11ec0c6a5ca44
55018 F20110216_AABTLR sachs_b_Page_074.pro
d733e174c38e246ed4e94c64649e42fa
743dab855f84e985445890d27ca015307cda3c97
55258 F20110216_AABTLS sachs_b_Page_076.pro
f132ec3bafa9a4efe99db9988b36dbd6
ea8d639f68fe1422a99783d6e09796abf7e9382f
54215 F20110216_AABTMG sachs_b_Page_091.pro
cf4d6a15eb628b59613330e7a6d4f5a5
4ad7ae11608c00c90d402fb3a3cda4c1ab01691e
55942 F20110216_AABTLT sachs_b_Page_077.pro
94f907b95486d3684377c8a26b6cf85e
269ab9c1e063f57ac48c3fe26458626779df0e93
21231 F20110216_AABTMH sachs_b_Page_093.pro
26b3cd576542cf861ad937bd7ef1a760
39b82348bb46dfa0eb6d80f2828970ae749fa9a2
54918 F20110216_AABTLU sachs_b_Page_078.pro
8b6ef2fe2008c6693c57bc6724271e35
50e91f920b3ce6a76bd166b02d404c31a2836716
78685 F20110216_AABTMI sachs_b_Page_094.pro
fb273199a64ef3cae24795c36bfb0c36
1ec97a2c0bebbbc2138862836b85fcbfbfd9e868
56583 F20110216_AABTLV sachs_b_Page_079.pro
4c426e295039818ffb0a4b93c7ba3eea
8d59c82598a57d723be7cf67361feabc07142a3e
12282 F20110216_AABTMJ sachs_b_Page_095.pro
cb5ab81d9d4f6ab78fa0b7e5b47025cf
c888f79cf0ac7fdb62c45b64dfff26791e7bfabe
56838 F20110216_AABTLW sachs_b_Page_080.pro
433ef63b34bf341b9630a052287759ec
820a621754261e14faca147cc9a359c73326e46d
6466 F20110216_AABTMK sachs_b_Page_096.pro
a926961961c1461e8e922cee43f2e60c
ef182a38b096f46765fac6b6bb2947e58753a13a
55908 F20110216_AABTLX sachs_b_Page_082.pro
b205ff48db9bd745ba9bcf322db90311
0ad4dfc2f71c58a04b360db5a63a833dcea86f9f
4286 F20110216_AABTNA sachs_b_Page_002.jpg
e6fd616e36613acd60713657f4457c46
bea074db6c81166b31bb10adc893f03106ff1983
58221 F20110216_AABTML sachs_b_Page_097.pro
dc375e47d417344f582d6bef4d8a96d0
97913cac06e25516e996825303705f59d8dbe4e8
52405 F20110216_AABTLY sachs_b_Page_083.pro
3fd487b3605b78c4f77f1cd65d2fb232
a363dc5bec99811b83f55c9ba3d26099492fd7f0
9548 F20110216_AABTNB sachs_b_Page_003.jpg
1b42d1cffbeae4cb975c302719556ccb
fb482b3955b3707f080e8a6de3a9f31e210255ec
57102 F20110216_AABTMM sachs_b_Page_098.pro
733e2b58deb2e5132d526de4f50e61ca
b936c3e34ac32886503bb62d366891ad3a16fd71
53126 F20110216_AABTLZ sachs_b_Page_084.pro
7fd0a61049d60fa153f9e67efc967909
f8cc11170ba6847da9d301ab18d6ff0bd0607d58
3523 F20110216_AABTNC sachs_b_Page_003.QC.jpg
bc39535fb2847210da55f3707d70f93c
3e33cb1ad56bfcf81c66b3d1eda802974c50b95b
68015 F20110216_AABTMN sachs_b_Page_099.pro
ef251f65b24384494164bf2e9a1a4aab
03aa46bd0bc8c0d7154268d7b97052b1fa2ad1ba
21018 F20110216_AABTND sachs_b_Page_004.jpg
0432609e8d8cc526a22e5c215ecdfcc2
04a2ff816c9423407682d1e650e1a19fc7cacac6
58465 F20110216_AABTMO sachs_b_Page_100.pro
9d7cf1c0f0227ecb6b907b3af3581e78
8c1a32bda9cc208f09afae270886980725aecd64
6377 F20110216_AABTNE sachs_b_Page_004.QC.jpg
2110c1648831c7fbafac8843b1043683
dbf0be1dd4532513e319700c1ae55e8eb8d70a33
63855 F20110216_AABTMP sachs_b_Page_101.pro
15d9a14665abfe757f4742337a9b5147
025c81ee80685a9bf96974a92d21435f550a816e
111564 F20110216_AABTNF sachs_b_Page_005.jpg
a2c891d950dec4b8dfba0c9a34950c5f
42f331ef53cc9f24875e2f28b96ff5b42ea99a89
65757 F20110216_AABTMQ sachs_b_Page_102.pro
85a782a061f51d2e6bdcc10658f134dc
1ec3c74f38ad51b0a63c5a9e589100371ddfcce5
23702 F20110216_AABTNG sachs_b_Page_005.QC.jpg
14c2c149d50db87ded9ec16c1b454cff
90b4eb448fff88056df73bdf647ce68f63855e32
60237 F20110216_AABTMR sachs_b_Page_103.pro
9e87cb179b5b5cc58b74f6d41609bef2
52d8ba4c1d7f174300e6637b79fd38849f564e83
62386 F20110216_AABTMS sachs_b_Page_104.pro
0e1fa189c1305bfc44aae830b216e12c
a0db182411c50e6836f1682e0bd68fc95a79c39c
74282 F20110216_AABTNH sachs_b_Page_006.jpg
3e6792af1e19051207240d7120116d0b
6efb7342d019cccf7fcc25949dba8fe13509bcd0
65141 F20110216_AABTMT sachs_b_Page_105.pro
2f4cde92fec1e2a2cb696722f29fd3db
8cc2b71c176d19eca48865c11df3531f71d43547
16258 F20110216_AABTNI sachs_b_Page_006.QC.jpg
b95026cbb14fd97eabfb1281cf6402a2
5804f07b51f9e86a7605c6eada2f0802f9330f54
59788 F20110216_AABTMU sachs_b_Page_106.pro
e3a9b0cf4f59b51345994cdd7dd40494
09232cb281639afbf77c089507d3302ad1981c85
71326 F20110216_AABTNJ sachs_b_Page_007.jpg
a823405fefdacd2bae0b44a6d9e8a513
2499a0f3991dd0e515129d297a9355d96e7aaf7e
64331 F20110216_AABTMV sachs_b_Page_107.pro
3e16b20910fed44a1f7698b45c216ea4
0e1855df6aa2e1fe6537c2c8136da3d9ba28a4c3
20499 F20110216_AABTNK sachs_b_Page_007.QC.jpg
48983e0354ea669b05b1b039e14659a0
403452b12f431bc13f5b2318e47188f118575cd8
34341 F20110216_AABTMW sachs_b_Page_108.pro
cf9d676bc6cb71e174a1097c9acfe44c
fcc9ce277ca2d77f27c2c521b367fa5b4f0a9f9b
103571 F20110216_AABTOA sachs_b_Page_017.jpg
92d2e6b914d2f6cd416bef61d395cfea
d142b5d22255c36e03c9d5d84253530c1805d93f
31362 F20110216_AABTNL sachs_b_Page_008.jpg
e9994b711054cf3318bf342330af2ca7
05354ccc22a723148005ab7454f543a7bc65eda4
25801 F20110216_AABTMX sachs_b_Page_109.pro
cd38e629de5182a3a6825f7f4f6cc6b6
162b686d6a8b530c0af0381ace11c1890fb441b0
34936 F20110216_AABTOB sachs_b_Page_017.QC.jpg
e209cf6b0f9e00e34d8af8199cf8c4b2
d92de39fce534aa4f91430552f4210fb3e37f814
9263 F20110216_AABTNM sachs_b_Page_008.QC.jpg
a3ce770baedf9653a55c64a77d975519
c37bd0de1492c121ef09b5c1f3abaffdd4f68575
30609 F20110216_AABTMY sachs_b_Page_001.jpg
6bffbe3206b873d94720f6af09a3e4f3
3f2a468eb22d50f52650a720f116555dc953962b
112635 F20110216_AABTOC sachs_b_Page_018.jpg
86bdb579c721ef6da44433cf6607f26c
0423954caecb44acfbb6819ae4aeca0ed49cc472
91388 F20110216_AABTNN sachs_b_Page_009.jpg
a8032137dd5111617a83ca5415fd91f0
b733795239b79078c352a7c897d4e94a96095be1
9042 F20110216_AABTMZ sachs_b_Page_001.QC.jpg
0bdc311e7909afe684965b43b7c448a3
354065eded8a57c822eeeae9f20b57017ccdc937
38158 F20110216_AABTOD sachs_b_Page_018.QC.jpg
a1842fb3aa1ebd0c8a933ee3c35ae713
174dd0fbf9b2330f64e07217ae16931ede0502de
65338 F20110216_AABTNO sachs_b_Page_010.jpg
8dd849de3738ead8d3cec69cce9476eb
e1a06cde3ee10bcb9a5aec02146a2ab35b371c5b
110022 F20110216_AABTOE sachs_b_Page_019.jpg
6db977afd96f6704ee2eb605a40c5d48
a1bbe67b5209f8dd109e8234485b258cc333dfba
21716 F20110216_AABTNP sachs_b_Page_010.QC.jpg
070adaa539d8003b27a9eb893975207d
166b59c1f760e4c53cc27d6f96b63e257c57cdc6
36190 F20110216_AABTOF sachs_b_Page_019.QC.jpg
431f7854a68cf33c0e197c6fb0b45618
c95ba0b3881d12c08aca7dc3e268ff48f85c0a3f
34847 F20110216_AABTNQ sachs_b_Page_011.QC.jpg
9acbacbf7e022e0080e3ab34bc2aaec9
237c5d767d482cdcd05361f15cf1711114b2d38c
110740 F20110216_AABTOG sachs_b_Page_020.jpg
ced651891c49dc26bc53a1afe27d14f3
a754614451317cd2b0764ebce1017d17ff864aed
109403 F20110216_AABTNR sachs_b_Page_012.jpg
c7a3ebf8dda422efb32571938df93c56
e880dd78bf7a4a0b6b52b7d9b92880e34d9b06d9
35258 F20110216_AABTNS sachs_b_Page_012.QC.jpg
62e4abfb606193777722b79aa08c9496
4f0125813edbbb29f9135b8f65644c227028c4c7
36954 F20110216_AABTOH sachs_b_Page_020.QC.jpg
8d3b226fb2c33855c8958b68201577d0
960aa9db0670caaeb6fadf174ad2eaa8115f8d3b
111265 F20110216_AABTNT sachs_b_Page_013.jpg
205ffbd2f49f5a3bf4a618f92d930151
fa748bdce1100d43e5663a0662f3105dfc449420
35844 F20110216_AABTNU sachs_b_Page_013.QC.jpg
3958f4bbfa4b5d1c6de590bd8f34b4a1
8b43d0c9b84a51c1b7b068aa9c4d2f504b43d68a
111143 F20110216_AABTOI sachs_b_Page_021.jpg
b389899cfff9072a3e08be344577321b
ece6ad9d0765bc7438eef2d14810a4fa72533532
120106 F20110216_AABTNV sachs_b_Page_014.jpg
ac556b1d140ea7a3591bbaba734519c9
d669546de8384614538ab1e51f6c4a7d61a1e98a
36073 F20110216_AABTOJ sachs_b_Page_021.QC.jpg
e0c43f155869bcf653a0ae5944418140
f3261fec8ebf3d3399d55de3b81e8ff1fb67a7cb
39787 F20110216_AABTNW sachs_b_Page_014.QC.jpg
d6849f99bf5b2e5db49e50485de6496e
63a1210b524de6afab6056cff7097d34f5948f07
106000 F20110216_AABTOK sachs_b_Page_022.jpg
b51710bfac89e6a841e1f045a563767f
ecc24ae3369e4bba79b8835b22d59f47860d5488
114781 F20110216_AABTNX sachs_b_Page_015.jpg
de8706bf15e60a56f4e90697b1a3e520
fac801294556ae5207180737281598779a0b948d
109960 F20110216_AABTPA sachs_b_Page_030.jpg
9c0f33939c559f5e2320dbd7e46d57df
75473979a2fb7031ede79c8d54b68987971e2148
34511 F20110216_AABTOL sachs_b_Page_022.QC.jpg
62ab4c590095c4e717784fb6737b78ae
645ded51417e2d59df25189f340f344073201227
112614 F20110216_AABTNY sachs_b_Page_016.jpg
17d6d3539d9cfa236ce5ee634a2ff512
1d37c2e74d11af746c7e7315c6d3053234f64c6c
36569 F20110216_AABTPB sachs_b_Page_030.QC.jpg
87a0cf885fae81e1f309be092f64d2d9
8d7f281ee9ac77c01c3ee22e48aa5a811286150e
103627 F20110216_AABTOM sachs_b_Page_023.jpg
03d2872654b5b3ae52a13c4e95db36a3
56daed7a0cffdf012af392678585c07249209684
36984 F20110216_AABTNZ sachs_b_Page_016.QC.jpg
b49de2b4101e7217133c38601590d0ae
c19a224dbbff8edeceb0f9aab21e5f29a6056406
110564 F20110216_AABTPC sachs_b_Page_031.jpg
f266b15ba61e22cc4e9874ba6abd7aa9
3f14707cae01dac2ad0710fcee8bdb7ee58ecf6e
34417 F20110216_AABTON sachs_b_Page_023.QC.jpg
081ab71846c8f2e8a96dbbacee1986ff
73b4fc0f1b4b2674574d96ce2045534df364bd23
36118 F20110216_AABTPD sachs_b_Page_031.QC.jpg
ed5d8b0685f18fc06e6ae252a72e50df
945c11a6f5018e12ce0bd679a9f15d6113653614
113627 F20110216_AABTOO sachs_b_Page_024.jpg
b46516ece32c6dd759182de57f0f52fe
8e82eeb7b529b832e40280dcdba2fa09f23635a8
106888 F20110216_AABTPE sachs_b_Page_032.jpg
5eb062ed54c6e636ca5117e805c4bdf1
62c9f1766829aefb5b4c52d9386ea6da5619560d
37309 F20110216_AABTOP sachs_b_Page_024.QC.jpg
d554e81e32766e9b47029d404e754fca
ec39ef260f6ca932eebc466a54233adec881d0d9
34969 F20110216_AABTPF sachs_b_Page_032.QC.jpg
3f30544791253a2283beac9eccf9aa53
93472f24eb2583b5da09aa32ecad4a0f5674c5be
112617 F20110216_AABTOQ sachs_b_Page_025.jpg
9985e434a585573386e70468b33d2423
05a36549c0e934315ef8e57b5704935c3e787724
105761 F20110216_AABTPG sachs_b_Page_033.jpg
e492feaafab5b4a4c31cec04f6b089b4
4245c35bcaa5b8a50d2710fe69707d74a44456d9
36719 F20110216_AABTOR sachs_b_Page_025.QC.jpg
dd66662c422db999c294e334ce569787
158a2c63a59c4a61e59bd9e905ce45f2018a6659
33737 F20110216_AABTPH sachs_b_Page_033.QC.jpg
3b462270a3ca12f6797bde7952610c05
55e29f66dfdba069bfc4e99f0ed7b00e3de3e1d8
110047 F20110216_AABTOS sachs_b_Page_026.jpg
c8b8f886937d226f4f374aa01604ada0
1300f848c51b50eab853fbd398c708e52f9eb08f
20008 F20110216_AABTPI sachs_b_Page_034.jpg
cba7644e1372eadb298daab479d0d1f5
afe6561205c28b711402ea500aac809ed443d4b9
35670 F20110216_AABTOT sachs_b_Page_026.QC.jpg
fe0d4dc14e2f905f7348936f81469508
cf45d7b565957db93034fa8fd5fca9cd808aca02
113681 F20110216_AABTOU sachs_b_Page_027.jpg
18209efa38880b85dd0912fc29409ebc
b8600a2728de4885b77448923963be8bf2e71952
7031 F20110216_AABTPJ sachs_b_Page_034.QC.jpg
af0a9420f6fbc0a67c167352793bb2f8
0c6b6d56529a8197f34501bd6d9c01900215b912
37353 F20110216_AABTOV sachs_b_Page_027.QC.jpg
c748591fa4db579545f7e57f32cd3614
ad24168613c6386efebaa98441e1766b89559040
93324 F20110216_AABTPK sachs_b_Page_035.jpg
28c707b7827fb5409dc0cc8a175aa60e
22e276b2dfeba4f3a1175f8e10e6b22ca16e173c
105128 F20110216_AABTOW sachs_b_Page_028.jpg
6bb618f01541155dce35316e0b412e72
ebc039e52ed717f08ec22de57cbfe09b811497ea
110172 F20110216_AABTQA sachs_b_Page_044.jpg
b4c1cf04b55284d6d3eb25f4da06ee1b
6eb472445a66865856279f7d251a64a1b2eaeaea
29663 F20110216_AABTPL sachs_b_Page_035.QC.jpg
8c67952b9e25fe3e2896cc2abc2ded01
0d78e5d11b063465d8296651b71a10d0d84e0a48
34709 F20110216_AABTOX sachs_b_Page_028.QC.jpg
4b9c5ecc8dbf90657192d77b9a88bf5a
3481aa5972224aaf71d1e3f83adfd8efe7795082
36542 F20110216_AABTQB sachs_b_Page_044.QC.jpg
41c7e4749f07490971c7863a95493e34
72dd970a1c38837d46526614cb762d1056cf11d7
90426 F20110216_AABTPM sachs_b_Page_036.jpg
c57b00648d406e9ab47273177c1a5fbf
1179665ae1f60ff558132ef52eb92af9d7176928
114840 F20110216_AABTOY sachs_b_Page_029.jpg
cefed2ebdf1cb923672722b767416085
522347689742419375943c55741286c80e9d30f7
112184 F20110216_AABTQC sachs_b_Page_045.jpg
0213aae0cfb9e3604b96e6d2a8cad18d
4986d943ab5443233664e1eaab18ab1056c5a3d3
29266 F20110216_AABTPN sachs_b_Page_036.QC.jpg
f6687da5628cf34d27039f240a55465e
bbe8c73e3ddb1c8dd199f1b9a1c9ee5e400dfe03
37350 F20110216_AABTOZ sachs_b_Page_029.QC.jpg
b927ec679ddbe212b7080c04cc033c50
d3328d4b9da60932b82c431ad3dc08d278174ac3
36541 F20110216_AABTQD sachs_b_Page_045.QC.jpg
1ebb13cf2f08b0f02e07878e3726c7a3
217eb70063ae9d9c1437da8e4a1e73b13072a2a6
105716 F20110216_AABTPO sachs_b_Page_037.jpg
9b555061e795737ec48f18e4269fdde2
73db03ef1aaa17c100fe322cf8ef4d127120714e
107307 F20110216_AABTQE sachs_b_Page_046.jpg
2c506577c7473efa803ac704fb419eaf
8aa1820dfdd9e627b7a9eeabe817298427a7c11a
34830 F20110216_AABTPP sachs_b_Page_037.QC.jpg
c35e488955710df22995d4b2690bb2e9
49bc598cf4291c94084e3921926113fce1f7c466
35615 F20110216_AABTQF sachs_b_Page_046.QC.jpg
0aec927e52452d66768e6a7e9978f845
ba5b3696fc204ccaf10a573dc1501401a053f6a2
113401 F20110216_AABTPQ sachs_b_Page_038.jpg
1a5888ff8ab8c90bf61a5263461b4fe5
fb815a360738012c44da36bd20603f0ad86aad5f
108528 F20110216_AABTQG sachs_b_Page_047.jpg
6dfc799dc8126d5b5742f186955802c9
1173fc5c102a938efe71fdf5e32557a3261a195f
37060 F20110216_AABTPR sachs_b_Page_038.QC.jpg
b6619b6ce68340a5673f13acf4769169
7d0742980a7296d9e623e83e42d51c29039459d0
35814 F20110216_AABTQH sachs_b_Page_047.QC.jpg
2462f65febfba516ce1cc704f02c9944
0f0f5f9def54c4d09e0fd742be3cdaaa33e58458
112783 F20110216_AABTPS sachs_b_Page_039.jpg
7585935006962251f2204111fcd33a74
800ceb3032f209a34573915778d0c76c04f85fff
107758 F20110216_AABTQI sachs_b_Page_048.jpg
9f5694469da02b9e0c1c8c72ce990f1e
9760ed2e772ec989585340791527acf776e42845
37508 F20110216_AABTPT sachs_b_Page_039.QC.jpg
2cde1788e48f91d854f66a3360aa0e47
4ebfdd328d9975939169eb18c85416d1799f34be
35610 F20110216_AABTQJ sachs_b_Page_048.QC.jpg
14559c7244d6885108204336c69c9987
628c313ee442a348d62f5c184c5e21d036fac57e
111471 F20110216_AABTPU sachs_b_Page_040.jpg
bbb686171454019d01e73b332ff6bf49
613aa6837b23e9dbf09c8566bfef3d51f52cfbdf
37303 F20110216_AABTPV sachs_b_Page_040.QC.jpg
9ae908f2c5006f9b128eb0ef98167df4
d53eeab8503f7864c6e97568fb34ffee628dde8e
43691 F20110216_AABTQK sachs_b_Page_049.jpg
e5e53d01b463d8e209e02a3c0dcb4f83
484f6e2d230f1c2a33ecb4a4ea40894f74992bc5
33540 F20110216_AABTPW sachs_b_Page_041.QC.jpg
8e6f2c62ffb917c6cbe3014213119abb
0112ae7645cea667ae496c1f879e80ee7d5bcc12
13513 F20110216_AABTQL sachs_b_Page_049.QC.jpg
bf535a13511b9cb4cceb7c8fcc945d10
7662eb3e1cfb3e47770711e682d96f42ea8170b3
101573 F20110216_AABTPX sachs_b_Page_042.jpg
1c770552a3a707781acff98451eb50b1
98b4d135ceb83f62de3dbd55ccb79402fa0fa8f9
37347 F20110216_AABTRA sachs_b_Page_057.QC.jpg
07aff03ba7a3d8e20e15156267a2e390
ff6048d08ca353212b4b5a7ca68243067e53303a
30658 F20110216_AABTQM sachs_b_Page_050.jpg
496b036d0e1fa02066ab0b9021c536af
ce013f3b4b160e799183afc324f1b87af6b13a11
34170 F20110216_AABTPY sachs_b_Page_042.QC.jpg
09e90a6a5174d5bacb28e442048f246a
209a7e7e3ba5ead46608797cefc9dd2f76ffa3a0
99576 F20110216_AABTRB sachs_b_Page_058.jpg
d7e145acbb7689dbe7922de0c060889f
59d34b60d5de7ec3cda528885d44ecd015642c14
9099 F20110216_AABTQN sachs_b_Page_050.QC.jpg
88acc535df3fd4a175b852d7d8e9c182
fcbfc19c2e737e44629106a1e5b4fc30f64c11b5
35289 F20110216_AABTPZ sachs_b_Page_043.QC.jpg
6443b46a9319ded6087c4cd022f07dd8
0a439d28abfb32a0b95ea0b9f1ea3b7fdb916254
33220 F20110216_AABTRC sachs_b_Page_058.QC.jpg
455d090d3f1bc2e2a502949e57f1187c
ab4a1970267c59bbf37d5e7618bcfaad3c3a04bd
96657 F20110216_AABTQO sachs_b_Page_051.jpg
b541a04cf2a6278075390d19fd1990b5
e253b682272300d5efcbf17ea9f17bb9c0ecc252
85278 F20110216_AABTRD sachs_b_Page_059.jpg
1580afc551cbd5686e26b1aa8a15567b
44df31618158810bd4972021fabc88c9f30acf68
31634 F20110216_AABTQP sachs_b_Page_051.QC.jpg
bc9dc0ed1e1513b05a5615d6c493af22
fafc1c0ac8c41b5d4949908274aeecdec4f2cbf3
28128 F20110216_AABTRE sachs_b_Page_059.QC.jpg
7b0644f85d6e1e41d48404a6560a57c6
cb3a5ea9fd3e45fa672cd1770682485e96c13e38
98538 F20110216_AABTQQ sachs_b_Page_052.jpg
8ec78385c74f35aa0ae4578604915432
048c5c79b3d45d1d83c47cadcd4207fc989db582
79083 F20110216_AABTRF sachs_b_Page_060.jpg
05181012e163c3501ebc2c4a36068ae6
a5ef4c3d15d0d3975be253f7ea17b284b79a0efe
107906 F20110216_AABTQR sachs_b_Page_053.jpg
4284e7ee9074567c49100be7f490afcb
c67c5619bdb943d457b7dc330978bb3bc7ec3215
25441 F20110216_AABTRG sachs_b_Page_060.QC.jpg
24beaaa3dabb6c53d4951b5144e700a6
10fb33d76c9002fd677f29654605251961f50d21
35593 F20110216_AABTQS sachs_b_Page_053.QC.jpg
838798ca62826668656076585474cb5d
90ed40eb09fd773fb3e4f6420805e6baa5e6bb50
60030 F20110216_AABTRH sachs_b_Page_061.jpg
f70e3b3987fb8dfdffef63888a965c44
d4993ae0d082d573cbe719e9d394736698ce833b
108929 F20110216_AABTQT sachs_b_Page_054.jpg
9c1e944b9ebf5c529bd625370f297086
7d7c75718c141a6996b33505351fefcf38a6219a
19819 F20110216_AABTRI sachs_b_Page_061.QC.jpg
184f64bef0840275234c24e140459d99
09ce272d1c99c94703544acd6b6c2443125c2bb7
35583 F20110216_AABTQU sachs_b_Page_054.QC.jpg
3d891fb0b6ed6c6c2daff1f48333fd77
5bb0a1c3617084915dc2e859dcfbd154e0bcf5a5
87281 F20110216_AABTRJ sachs_b_Page_062.jpg
eaa2491e110d2cd5bf23242d574ba641
9a10821d5f04ffd5142d6cc5fcbe8448a4d46530
110460 F20110216_AABTQV sachs_b_Page_055.jpg
cd3d9c64dc77f0f17e5b71858df254c8
be4457f09a729889142f074d0f2ecea4daf4a62d
27098 F20110216_AABTRK sachs_b_Page_062.QC.jpg
cd0348e96874b895d36ef6effa5f9a10
84f236346ed08f0e3913f97eee648aef52834c81
36457 F20110216_AABTQW sachs_b_Page_055.QC.jpg
d630e457bd7177c928215c70f3d04ca7
17818d4ca77300b0f29ec7005e629a26021f6e8c
111825 F20110216_AABTQX sachs_b_Page_056.jpg
873fbee8d61e027c8696089114682bec
c8659b03d55381bbe6f151bf65ba0e34c56ab697
111003 F20110216_AABTSA sachs_b_Page_071.jpg
738909b01700f3fb7dc1a9a0bb238647
ca35441b10055814693cd7e6fec49edd253f32a9
51686 F20110216_AABTRL sachs_b_Page_063.jpg
a5f41e46eb1829926a1e5e86c35ccc7a
6e992c13416f929c13a0ff33bb700a110fc966da
38195 F20110216_AABTQY sachs_b_Page_056.QC.jpg
b92e4332d2e5119f691ce10ed7624378
04305622dc2b3c371058f8afb996ad4b1341f228
36200 F20110216_AABTSB sachs_b_Page_071.QC.jpg
3f5c8ad231061c8ee4bd25c064c8bb66
ba3d56c1cddf8508e44001b9402341a1fc0b68f3
17937 F20110216_AABTRM sachs_b_Page_063.QC.jpg
07d3837c6ca3d2aa779933fff7f5cf07
a5089339c0b05850bf1f998e37cfec960990bd6a
113453 F20110216_AABTQZ sachs_b_Page_057.jpg
43f1cccfa116989a84341ba12673b740
4d018f5958b4674a9b292b8406a262e2ed62cb24
107316 F20110216_AABTSC sachs_b_Page_072.jpg
d719c41838746d1ff35ef307f4b4b44a
188d70be02e9289bb0979f3ae85035b77f8f991f
19667 F20110216_AABTRN sachs_b_Page_064.jpg
5bf022aea10c5e80ce2306e63ad9c706
1801c805980211ef3f80f9202b4e3ecaa3e88e60
34958 F20110216_AABTSD sachs_b_Page_072.QC.jpg
5895f19f75d2c6a970299826c4a1098a
f988d5338079241a8908709251a015430eed1251
41066 F20110216_AABTRO sachs_b_Page_065.jpg
96c52718949491a577a70369e4e7d8ef
a2ae6a1262910ef2e65209364319f1df5ad14911
108690 F20110216_AABTSE sachs_b_Page_073.jpg
f8619f4729c32a62814a2cba737b68b3
628797420189a5eda3c6a43b1bd66f2f4824bade
16763 F20110216_AABTRP sachs_b_Page_065.QC.jpg
9aaee8510e81390d030e0189cc1ac22a
b2fc7de400afc1ef1f329f2996f486f1fca1042e
109672 F20110216_AABTSF sachs_b_Page_074.jpg
4bf3144d8b6e53bbc0f652d0927c04c7
9fe8dc833c7b87c01932d20eaabbd396899fbe3e
22876 F20110216_AABTRQ sachs_b_Page_066.jpg
e53d11471d83b9bbc124df62d34c88c8
801369b4e89525ed3fa224f35ec38d5e6fe19048
35659 F20110216_AABTSG sachs_b_Page_074.QC.jpg
7431b7637c25471bfc8441b32a63e968
2a1c6391540cd8d28b95b91a9ecef9ca2c88395f
9115 F20110216_AABTRR sachs_b_Page_066.QC.jpg
f93ef0ccdce7df344eaf9e2255972e9d
f6ad9de4cf5b97f9f2b8e1dd49e248b8b203568b
113871 F20110216_AABTSH sachs_b_Page_075.jpg
b05c88fa7a3699fbefa0055fdfee952f
9784f58d409ab5c314a7bbce56ae19b4bda4848c
105203 F20110216_AABTRS sachs_b_Page_067.jpg
74d00b85801965b6b59f98ba1fc14901
c00fbcc394e7c3bcbf200d4c7af62714001576e6
37122 F20110216_AABTSI sachs_b_Page_075.QC.jpg
2fda6b96c25b6f5ffe29d8ce4e46513b
feef59c84059865d3a56f78f6178ba3992680c2b
34113 F20110216_AABTRT sachs_b_Page_067.QC.jpg
f492ea43bd0516afee085374510e9472
06430af0d467316095a9f68865a42c888cc1e967
110230 F20110216_AABTSJ sachs_b_Page_076.jpg
3bc593d9a0eb09e79846af7228ea74f8
7313e728e434d717e9c7c2d36092bb12055d7296
113986 F20110216_AABTRU sachs_b_Page_068.jpg
c3c43f42ffe17e44ac418e4eb4e31f31
f2f7eb1b1bedc9266180946db4e1012653c2ed0e
111357 F20110216_AABTSK sachs_b_Page_077.jpg
771ce5534ae02d969335f92243d16d40
4ed15d9ca2e2b8b16fc0083f44abf50584dd6540
37585 F20110216_AABTRV sachs_b_Page_068.QC.jpg
526cb60069cb374ce4e35290558d845a
c4cfe5cc734aebfcb342e5563cdf49985d5d527e
36201 F20110216_AABTSL sachs_b_Page_077.QC.jpg
28509990d198b43a0336c9449b95fae9
9fb796868688cf05e0aa863c3dd8debe92154cd9
106192 F20110216_AABTRW sachs_b_Page_069.jpg
786bf2d289079bd20c67cb33b1f5c31b
448a28e6fcc1b28ae674ba5e5cab560e214853a8
39145 F20110216_AABTTA sachs_b_Page_086.QC.jpg
0d993d89d75d7e30ec48a72ef154ba14
0b622b779d25fb285a29bd20063dcd4d67671db0
34517 F20110216_AABTRX sachs_b_Page_069.QC.jpg
f8255669498c4ec23eebe0693b320aeb
4ef621b5e2686342ae98806d1c482a3aafd77f8e
112270 F20110216_AABTTB sachs_b_Page_087.jpg
532aaf841066fbb26d6d70cbcdd6e574
0fe6e9c29b475231717b75c0e79f30acc83473df
109919 F20110216_AABTSM sachs_b_Page_078.jpg
0cfb2cfef1465c4e56a04d571c4e0e63
75bba46350a051fbfcbef5b4e24330f2fc50b378
107606 F20110216_AABTRY sachs_b_Page_070.jpg
b8c87e5797d56519accc8c4557ca731c
b867db0baa5422f12c165283f8b117518d4eee99
36972 F20110216_AABTTC sachs_b_Page_087.QC.jpg
9346392f2680a6304a2cef98c816d5d3
1d5b89933d7ef917cbb61d7e56f407ff297c3b66
36208 F20110216_AABTSN sachs_b_Page_078.QC.jpg
f5593895bec991c1e92bf9f7fa4792d5
ef29c84f0db2d6f41ac6774b9d22d2c44c9c8150
35454 F20110216_AABTRZ sachs_b_Page_070.QC.jpg
d9304f5bb8e0e1f1e5c7352c2af544df
65993f8c34094149d3b4a2d42b32385627ebb178
112516 F20110216_AABTTD sachs_b_Page_088.jpg
8e0ce3eb44373941c80c62b630788f77
4e80708cdb9274cb30364a4dbdb0e7670fd1cb57
111755 F20110216_AABTSO sachs_b_Page_079.jpg
9d10886d0bc6a894f6aff220d0319fad
462d5372bb1817997214565fa415206c4404b12f
37189 F20110216_AABTSP sachs_b_Page_079.QC.jpg
eae480319208471c0150e24830ea4ce1
2a244cf0b9f5e5b1098e477806a9fb6d4eb8ae59
36862 F20110216_AABTTE sachs_b_Page_088.QC.jpg
b23c37424e9a5ea7f1424f40a003927b
4f79a4e894544b1aa31fa84b573f1967aa6a5672
111194 F20110216_AABTSQ sachs_b_Page_080.jpg
2c14a6e86de18510e163410243811aa8
a9d08863b7b1f37c7aefc344e63120c0057958b2
106129 F20110216_AABTTF sachs_b_Page_089.jpg
6429a1d27a878fe88314a387bbcfcf5d
f088d80f53f6d4c8362b7b2b6fb2d27de7615502
37615 F20110216_AABTSR sachs_b_Page_080.QC.jpg
e0da4a015900ad105c25bafcf9745b79
1731cfad53f4494b52aaa3741f6edaca51b1a27c
34327 F20110216_AABTTG sachs_b_Page_089.QC.jpg
24827b9bd9ba5e1bab9eb962c7b7bc0f
c4f8b7860e8d2b0b907d9b439c981453c00e6edb
114907 F20110216_AABTSS sachs_b_Page_081.jpg
9cfb1d5ca664652885057b7ab3274bf0
4c46e4306f9932171d74b8f30e91cc4124ca861e
108518 F20110216_AABTTH sachs_b_Page_090.jpg
2048086a683065560652ec8d6d0d2fd4
8867618252065d069170d47f0d08d46d9093c09f
113423 F20110216_AABTST sachs_b_Page_082.jpg
5e252e69f799bf1dda03eb243743c8f5
0a596a8684d8ecb26a3c38b722d05dadf52d161a
36044 F20110216_AABTTI sachs_b_Page_090.QC.jpg
62170a833079c17520d9b996bea4d414
3abf288c4fdaa90e8d91ff5658933c17fc26a72a
105579 F20110216_AABTSU sachs_b_Page_083.jpg
6140198547bc82a251137aaed8c57cb5
b58606aed17aefbc98a7c5e4423bbde68c4305c6
108165 F20110216_AABTTJ sachs_b_Page_091.jpg
da8171b6269fdc1997276728bc401429
1e5ff7f11d65118826d50280272d045c1a5de1cc
35191 F20110216_AABTSV sachs_b_Page_083.QC.jpg
fc69da608167559a71849df542290eee
8fbd21c451260034840af09c9b47a855d63d7f6a
35394 F20110216_AABTTK sachs_b_Page_091.QC.jpg
c1796eb003d3c28623fa09f62b61456e
bc53807c52737f22b5808ab40ae188f71d93a26a
106733 F20110216_AABTSW sachs_b_Page_084.jpg
c8148ac67a01b87cb43c83181cc3181f
1b48c064f500f0699f81ead7b7ea1d2c8af0f683
113761 F20110216_AABTTL sachs_b_Page_092.jpg
f5b42ba259338f65b1adeea79f24be6e
82a3c615bfa5ec55fa7593dc57d3876848432cf0
35140 F20110216_AABTSX sachs_b_Page_084.QC.jpg
4ee9c6b2f74e855e17700c796ffce799
bb2694c81b942fd070c3b19ce67a021d448fc6cf
118751 F20110216_AABTUA sachs_b_Page_100.jpg
f37ea56538e36e3aa9940cdbd655d6e8
656a370ddb7cf54f4c0eb1c94c50c7132ad96ec8
37301 F20110216_AABTTM sachs_b_Page_092.QC.jpg
93fbe81489702b61c3681bf5d317a54e
c44ad973e9fcd8cd3f69d6495f3bd65d002a76ef
113383 F20110216_AABTSY sachs_b_Page_085.jpg
504387c9c7d1a4f42ee7eeddf68493b0
79de85f44a335557298b179b975c62b8d6415a23
34235 F20110216_AABTUB sachs_b_Page_100.QC.jpg
1e7343919e95798c8b70512fa426aee3
6e01a725acd38d6580fa4a4b5aba9f391b5125b1
36803 F20110216_AABTSZ sachs_b_Page_085.QC.jpg
882d5fdd9332c7ab01e39c9567b26aa0
652b3d506cd145d8e0396533ff9c4946872d8841
35927 F20110216_AABTUC sachs_b_Page_101.QC.jpg
bee4d0ecef8d35b4b129d6f2582510b4
c5c7150e5dd04efd10b5d004dcde8dfa97d9aa6d
45112 F20110216_AABTTN sachs_b_Page_093.jpg
a35c4f1ae4fae6b1a580a0425dab1bd5
cf7dbc22752ece979e92d23ceb5844ef65f13605
129030 F20110216_AABTUD sachs_b_Page_102.jpg
7dfe607f164388d1bc81537a996386b4
1a56ce5762bf85877eb2d98460805c733070b12a
14904 F20110216_AABTTO sachs_b_Page_093.QC.jpg
12a75f5f4d52ddfdacd8d2b6ee03f3de
dafa2a97df0c1acb94e1fbc19c8b2cb5e55a8d32
F20110216_AABTUE sachs_b_Page_102.QC.jpg
f0651f440dfcea81da2e74c70ee93829
2b5f83bc061678e2f5972d427e47ba0cc4735450
39486 F20110216_AABTTP sachs_b_Page_094.QC.jpg
d23c90d8c8062d3bcbac02a7c28276b5
b0519536cd40e624c1448d45c5d333d22f523ec5
120097 F20110216_AABTUF sachs_b_Page_103.jpg
0a7c5abfbe0a0c108cdbe106be9e3e40
50bd175c0ef8e12ee65bb3ca8ee72e28ddcfe4d3
35179 F20110216_AABTTQ sachs_b_Page_095.jpg
cbe762830c0348f757f2b2141a3e990a
e1301f8c962b05bb455e19a009a138f9eb91cb4f
7939 F20110216_AABUAA sachs_b_Page_051thm.jpg
298a71762c31289f4014ddac32679698
20091e95a2f2bfde73f041753bec449e6c12a69c
34336 F20110216_AABTUG sachs_b_Page_103.QC.jpg
c4f8c00eef88f850b05f5b1b2423743e
f77ae952e239e1fea3e93e0473d57ce92983fec4
10495 F20110216_AABTTR sachs_b_Page_095.QC.jpg
429a4e0335a664f62390872ed364cf74
56f27cd0e7e0197ff7ef2328fac5d8abe3264dd2
118098 F20110216_AABTUH sachs_b_Page_104.jpg
2ae302012b0f497d7e848227700ae262
c2c23f286808844da2cb0ff6d3c7c36d508555ec
22735 F20110216_AABTTS sachs_b_Page_096.jpg
88469c74465d732f9c4415b6507798df
c37c28d115185c51fcab4edb572496fee8cdd825
7814 F20110216_AABUAB sachs_b_Page_052thm.jpg
db5834a13cb6525251533dc024b13312
a78ac6d7f1d510e5f81ab0c5daf1c94aae296512
34606 F20110216_AABTUI sachs_b_Page_104.QC.jpg
2c3b0fe8ba58311da4998ae4eb39f6e0
8d13f595a58e0e9830146114fcc4d456e29fbc65
7992 F20110216_AABTTT sachs_b_Page_096.QC.jpg
40763218037fbd54d1c30657e51125eb
7144bd98a7ef8794d6347931808704ba767a78f2
8652 F20110216_AABUAC sachs_b_Page_054thm.jpg
cd10caf52ebbe4004f9ffd86a46013da
47b746526a5d169fd1ecf30e5c0820f98ec25bc2
36227 F20110216_AABTUJ sachs_b_Page_105.QC.jpg
1a26d73a8e4ade5b7febfabb69881107
59c58459cd458769e825110867d05ec74c21d73b
114689 F20110216_AABTTU sachs_b_Page_097.jpg
22378a4cda0098459c08e05070cca677
7e08fd9cbd2fd032e720d4923bee8321c36e6278
9040 F20110216_AABUAD sachs_b_Page_055thm.jpg
61c4de58846b765c620cbf18b279b124
df0dbb959fa34b5adf693f054d1bb6345ac92511
117318 F20110216_AABTUK sachs_b_Page_106.jpg
d84fd6d3b34e1421f4c7f5ac7dea404d
e5ee319de21d3d60a5d0d394511486cdfa609874
33018 F20110216_AABTTV sachs_b_Page_097.QC.jpg
9d619510e4c6b8aafa82b2dc17df73ed
626f8fa58bf95a79e25380f185600c8e63dc481e
8883 F20110216_AABUAE sachs_b_Page_056thm.jpg
40e80faeb9be0bac1abb6310c0f6d883
f8d9cb7ab72219e3ae62c472514db0fb8f2cdb66
33509 F20110216_AABTUL sachs_b_Page_106.QC.jpg
b39e086db195f0d465f4466e4e469445
1e76e717cad8774b1c7444ae27f8e2e131cae202
113972 F20110216_AABTTW sachs_b_Page_098.jpg
304122f55265dff2bc24a662181148c0
f8e3e392ec365d9dca8b5348fee0099c241108a6
9148 F20110216_AABUAF sachs_b_Page_057thm.jpg
bff33e3b3326c0a80a79d291cab05aa8
95a196faf71b806865d4e5560a85da7a46dec63f
69961 F20110216_AABTVA sachs_b_Page_010.jp2
4a902dcbfbdf0b57e598ad4c904cbf69
791055d5096d0fc9e0589df02f578d6176a8d8a8
125218 F20110216_AABTUM sachs_b_Page_107.jpg
5f294ec7ca75c110f7c70eeac33b0b40
573a2b903dfe489e16057e5de5ca3caf08bb7545
33145 F20110216_AABTTX sachs_b_Page_098.QC.jpg
c4c83e5fed9ebbac291d7ad3bbb700e0
f057aaef4039caada6b1799d9778ba48ba36d5f1
8504 F20110216_AABUAG sachs_b_Page_058thm.jpg
38be0df0fe6e2415127c6d3c2814098b
1daecae1236e24d1c28795accb2ce35009d795d3
113526 F20110216_AABTVB sachs_b_Page_011.jp2
54d107dc1e0686d09ce5006c822351ca
ca1b599455c02f4cf00161f1242b1e486ac8b76f
35445 F20110216_AABTUN sachs_b_Page_107.QC.jpg
e8e248b644bb18f5cb321007f370f6a8
93f91be9b6d43a4c7fd25b144bb80e379917b203
132515 F20110216_AABTTY sachs_b_Page_099.jpg
fcaff8a92d79d053a3ea60b739a20a10
eeb1e8957099d1cd3af8eb92b771c13425d8f068
6841 F20110216_AABUAH sachs_b_Page_059thm.jpg
a28835193ab29a8c0fb6f06a1c6b13bd
0218172624a9ac91ffeb8c07b8ee91a32b9dd4b4
116905 F20110216_AABTVC sachs_b_Page_012.jp2
d7fcb0d533cb46f866be13ceddd1f54a
2a95c8c02295d2e6897df2b826f4aa274595faf9
37153 F20110216_AABTTZ sachs_b_Page_099.QC.jpg
177da6b5499bccefdf502afdbe718342
d50c7d6f41ab5c51f6a65ba9e8348840182d1da1
6758 F20110216_AABUAI sachs_b_Page_060thm.jpg
c5243439be30f178ed172220994a6a24
88c503af6a9430b722c9fefe4e209b47595a2a53
118605 F20110216_AABTVD sachs_b_Page_013.jp2
08b0a36071880053bd252ec226a1ea6a
477681abeba197077258ac8957ed4e363881e23f
71727 F20110216_AABTUO sachs_b_Page_108.jpg
10711d51ddfdc39d77c61ee8319d5863
adfd4e45722a0d022fd48c1b9bf63da615bb26f8
5696 F20110216_AABUAJ sachs_b_Page_061thm.jpg
28dc126c6e16f5029dba6f205493621f
f95cde2923d6a838cbd84d86fc722055bbc018d3
124630 F20110216_AABTVE sachs_b_Page_014.jp2
40528de82a5336c5710e26a9b38d8851
f3edfa23607f788fdcca58ad81d55666c49272b1
21120 F20110216_AABTUP sachs_b_Page_108.QC.jpg
ae34748225fb50c7b655a45a9e21eea6
aaa23b764d05051618237b7e9e0d7ada3db22c8f
7315 F20110216_AABUAK sachs_b_Page_062thm.jpg
db2b6540dd1e9e6d5d73ff1a7ca82867
db2df68e1827cbe949bd0ed1f40a43cc652e9ff2
122015 F20110216_AABTVF sachs_b_Page_015.jp2
a68dcce8a217b3dd8759479d2dd18ded
da4046ed5694578dcc4b23fa2980fe0108748dca
57210 F20110216_AABTUQ sachs_b_Page_109.jpg
42584682d424c46a215470e772964e1f
60340cdecece40355dddc9118c95c76e602400ed
5607 F20110216_AABUAL sachs_b_Page_063thm.jpg
5c1d2c43effb25eb50755ba0e910cc68
46949ad25ea20aed1e24b2d14b7703ec7fd5aee8
111526 F20110216_AABTVG sachs_b_Page_017.jp2
a1253ebba503bb458a8788209b8c3f3b
264be4c580d32d29a9e6dd62e9c0c207b7eb0cc2
18750 F20110216_AABTUR sachs_b_Page_109.QC.jpg
17a50549bff8f05dd2cbb832152f03af
3a090281f13e2ea7f3d9dedf64d8c40a8ff7e18a
9001 F20110216_AABUBA sachs_b_Page_079thm.jpg
df7e2caeb7fc46629e302c21307a38b3
fd3048dfa6ccc26391087e307c2dddb37425bda0
2476 F20110216_AABUAM sachs_b_Page_064thm.jpg
46760461219ac9c8552fd9527ccd2ce8
34d8ed7b33d6bdcba5f06e8ec6d540f0c7ea3f99
121024 F20110216_AABTVH sachs_b_Page_018.jp2
cfdf34e7c75f559ef41bf15c89731e58
0d99295824c5d029b100a879c320055caa66535c
5791 F20110216_AABTUS sachs_b_Page_002.jp2
29643d7d71fd40ff90487cc9a5b8eec4
af8189ee59c367012d719bdee605f086817291c7
8971 F20110216_AABUBB sachs_b_Page_080thm.jpg
8370e621d12f7076cd48c63cc8758d32
37e7243d44cbc2192d798b4c266f735f189c54cf
6513 F20110216_AABUAN sachs_b_Page_065thm.jpg
6197685c3948e897d93f17e900768a5b
329bdc7c008eb79a4863ef1610b0863fe35bf13b
117673 F20110216_AABTVI sachs_b_Page_019.jp2
64d28dc09966f1cdea2f622abef8c7ae
3d56fc5fbe97aeb9e2294b9b4cf49f13a7e04ed4
11243 F20110216_AABTUT sachs_b_Page_003.jp2
fbb9df3f238687bab0bf3ca82bc8c3fe
9813611e476a849dfa9226cfaba7d3dc6ac6e58a
9245 F20110216_AABUBC sachs_b_Page_081thm.jpg
932d7efa802f65ea339daab21d735541
8b4241563932b17d7ad8dcf10c242a565bab4671
3672 F20110216_AABUAO sachs_b_Page_066thm.jpg
3dd9e13cf254d982c017c6e4fedc1943
4e5ef502bd802f80e61d30df416629d822f23337
118122 F20110216_AABTVJ sachs_b_Page_020.jp2
140ffcaeed9c17b7ed997b97fb042086
89429a2807da5d3bff9ff6c8a7570847157ded4f
21155 F20110216_AABTUU sachs_b_Page_004.jp2
af1fbf4a700c9836990d4aa3417b8eab
8e5989efe62fb8a5728708436b030fe7d4fc174a
9091 F20110216_AABUBD sachs_b_Page_082thm.jpg
5c19833ba808710746ec9ecde7275924
6960add6f014d396ff62cc9ba07f840e2048deb3
8345 F20110216_AABUAP sachs_b_Page_067thm.jpg
39b15793a9edc71196fb2186e82dd1e0
392b788acf9191875aa1330fd1afd1eb2174f6fb
112963 F20110216_AABTVK sachs_b_Page_022.jp2
f81eac80fe89cfa9662a9841e0cfc954
94a73e636c198464dd2469dbd79cfd87c0b67f78
1051957 F20110216_AABTUV sachs_b_Page_005.jp2
e581fa264df5c0bb65abf8110bc4f7f2
a48d96a0cea93ec1d8ac25052ac1ebde7d2143dd
8848 F20110216_AABUBE sachs_b_Page_083thm.jpg
83521af4fd3651ae5422987ffc408d80
764cfd5e2d9e89c7408bdd9a8e2a1b6429ac591f
9261 F20110216_AABUAQ sachs_b_Page_068thm.jpg
6e29829d7f518aaac443f80389184698
002db71d8cb8d238dd241fdf06d22ceac028c289
113622 F20110216_AABTVL sachs_b_Page_023.jp2
f87905e5b8d39adae7b37a64313f7f10
1d56407a958aca5d876e38ff026efc3abe83749f
1051905 F20110216_AABTUW sachs_b_Page_006.jp2
1349ee6a9e95c26ddaf69b222b79d7f0
b6be0cd7092888b55c3e41788d0c57f3a89df326
8464 F20110216_AABUBF sachs_b_Page_084thm.jpg
dbbd41fdafcaa47303ec7dcafe3e319d
33d69671b8a6249711bb4c7e4b48cd62e123b4a2
8471 F20110216_AABUAR sachs_b_Page_069thm.jpg
b0d5a37974708ba324cd5719db6a95b5
1d3edf8490e2b74a77b647dca4c9bd2d7edb7f8c
121143 F20110216_AABTVM sachs_b_Page_024.jp2
f30bbc7a410aa5314d6b047532404982
7a5dedefda950b6e71746846b6a467ff59661859
1051986 F20110216_AABTUX sachs_b_Page_007.jp2
a49e337efb33d3a5021954dccf7088c4
ba4ecc893ec341983e46271d067e7a7da2797a52
F20110216_AABUBG sachs_b_Page_085thm.jpg
892b57d21c36f34f9896abfc5c520cda
8640f7f4465ecdbd0200126dd27a1621f3f737f5
118341 F20110216_AABTWA sachs_b_Page_038.jp2
abe09343d522a1a0c92129adb879fc9c
e4642775678ad5954b4c73b488187d958b807b51
8648 F20110216_AABUAS sachs_b_Page_070thm.jpg
fc1dcd31a0d9b765375fd12348818a48
920b89cc26dd994fd53d940e711a75b7fe535fec
118411 F20110216_AABTVN sachs_b_Page_025.jp2
7829a18866455960370c5d0fb21b6205
3a7629fa9a52bcca38dc89b49fde7cbc1ffa6670
538624 F20110216_AABTUY sachs_b_Page_008.jp2
11b8a65c6f669ee2304d51c0ff69349d
b4de701589878715e22d12dc9378bec9143f1a23
9190 F20110216_AABUBH sachs_b_Page_086thm.jpg
742cb669570e34383ddc4b158727aab8
8d3a794756310cd344ce78b594e475928144a715
117617 F20110216_AABTWB sachs_b_Page_039.jp2
e77da30453b7d4ce0b7b7f7961cf895a
5fe70042c7fe9653fed2a0bbb4aa31a75378e5f6
9055 F20110216_AABUAT sachs_b_Page_071thm.jpg
689ab497c0af40a42c135c9cd55bb446
3c34d6ee7765cd4987bf3b8e7dde1d2f81a87ca2
115722 F20110216_AABTVO sachs_b_Page_026.jp2
dda5ccb3a20e88c8af2006e51adce7a3
6deedac99cc4703f7c86e50d82fd0ab48db627ad
94097 F20110216_AABTUZ sachs_b_Page_009.jp2
fa074a117b09cc82d1bc3a2ff556be4f
5374ddced389b315949e94143cf029edea100c63
9171 F20110216_AABUBI sachs_b_Page_088thm.jpg
f4647f04857545742e14a0950c2f8d3d
d30b0eef566e117e214b9851b73a839a9a6c4824
118206 F20110216_AABTWC sachs_b_Page_040.jp2
568652a5ac4f254fb55cab3efe209983
ce3d5fe681a329b534c057f606273dc31b8bf389
8657 F20110216_AABUAU sachs_b_Page_072thm.jpg
b7ea8ce43e6abc17e5448a489c7997a8
5fa99e2cb3a53c4fd863583cf3f6e27494554188
8512 F20110216_AABUBJ sachs_b_Page_089thm.jpg
3272eaf2f8aa23899d107ba7262c91e8
e93e3a63b148a09733b35aac036dc54c05c2bafa
112318 F20110216_AABTWD sachs_b_Page_041.jp2
a057097ea646079ee7d8c211d638bb3f
31c3b447ce414eb7def06633816ddf71d56559cf
9047 F20110216_AABUAV sachs_b_Page_073thm.jpg
01c74debb200a1d69c9062f0d79e2cff
d9ab7eac723f77efcbe56be2bf945b81526089d1
120864 F20110216_AABTVP sachs_b_Page_027.jp2
fc8edbdcc9a049d3c271d3c7a848da14
a3ded6bb8bcbb597a142349d8223e9be23d9ecb0
8914 F20110216_AABUBK sachs_b_Page_090thm.jpg
3ca785565e4f57890e00f1b27cd50b16
6fc9b10e11e75c68418d01def400b39172f3c807
112480 F20110216_AABTWE sachs_b_Page_043.jp2
46e331bed78dab4fc77ccddb9f10da30
11ca3f93d7c86a0deae42ec65b782b1ed15952d4
8885 F20110216_AABUAW sachs_b_Page_074thm.jpg
c66fc84330d891d7e4caf5f5e606ed04
f9971477e20c27610a86bf39a2334598960f77f2
113331 F20110216_AABTVQ sachs_b_Page_028.jp2
2f6a0304d66d7dee2358436bd742e0b6
0b00659881a1a2c0645c95eae7ce17897caf4986
8893 F20110216_AABUBL sachs_b_Page_091thm.jpg
ea0631fd3de15cdc803029fd4d6e5f5a
818d6b9db88aaaa4d4ff86b11269cbfa7dbdd29f
117262 F20110216_AABTWF sachs_b_Page_044.jp2
10b7dc7e2d91421825bec171b7a9a863
79c1d3cd9fb9870b5133c481af63e83ab2aabeea
9203 F20110216_AABUAX sachs_b_Page_075thm.jpg
723c193f4ad1e5a6cea71bf6ff4c25d1
3922835bf3758adedd1f963981514f1f4f53e770
121750 F20110216_AABTVR sachs_b_Page_029.jp2
38175a628df2d81afac3b717f25fea73
a6c9df55f98ac6e2c0bfaf1e469b6bb3e1277a04
5402 F20110216_AABUCA sachs_b_Page_108thm.jpg
de70241949ddc2a7ed6758796d4c3ed5
15eaca7e45e31105246085c45105e9b4ebda75e6
9049 F20110216_AABUBM sachs_b_Page_092thm.jpg
583592e8726609fcd8ba70b757578d1b
be5578fc29d26ef8bad1150d0582d493fe5f8941
121081 F20110216_AABTWG sachs_b_Page_045.jp2
28774550ee6f599abfcf3515015cad2a
0b309b04b5037d2f008eea8d0d45e891ac8b7195
F20110216_AABUAY sachs_b_Page_076thm.jpg
85b7ab7eb18b1f5e3a1f55db0ec9d05f
cb83a3b91d7a28f85a25a6cf0bd5ea5c93d8c458
118044 F20110216_AABTVS sachs_b_Page_030.jp2
661bdf0cdc9a782b822d7465dc72fa4e
2eb9c95758f9635ad0b9b68dd5940414371eb7f9
4357 F20110216_AABUCB sachs_b_Page_109thm.jpg
9ad318d5eeb3f59b12a6fa6e09435cca
0cb30af95c34decab35d3167d135bddf1938067b
3804 F20110216_AABUBN sachs_b_Page_093thm.jpg
4a18576cd9c05756086d4fa631712e67
e17994493f7ba9377f4134dc567ca642b4fa8bd8
114073 F20110216_AABTWH sachs_b_Page_046.jp2
0be145767eb5f40ce75bf6f299b19d29
7eb17816ce5c0b70face186ef61b70456117ec6b
120009 F20110216_AABTVT sachs_b_Page_031.jp2
50493b8ec5595bd90b6fec9cac74f620
fd52f79ade569a60566d1f12ffdf186b33806627
128943 F20110216_AABUCC UFE0024780_00001.mets FULL
64112fac7808714bdbf0872092c181ff
8f02d966bcca454031d66224b361435ab0d5738c
10103 F20110216_AABUBO sachs_b_Page_094thm.jpg
372a05ca090caf98a5024dbc17330ada
83edb7c34730c7d07275ab1ec59846b4cd7298ae
114749 F20110216_AABTWI sachs_b_Page_047.jp2
556819cfe9abf122d46e806157e51db7
e103330e63e58b6584cd7ff8d42dabd7dd4a8039
9117 F20110216_AABUAZ sachs_b_Page_077thm.jpg
d636aaeb0346b56ec5e2c5f45055cbd4
87c3f9c242617ecfd5bf21f229d31a59a0376c91
115633 F20110216_AABTVU sachs_b_Page_032.jp2
12b5e5ce96fe0f9afe3cd35ed7320767
e5bbc9cbb9eedb0b62f1db0dcbd91ff6c91d5e6d
2891 F20110216_AABUBP sachs_b_Page_095thm.jpg
5c8ec35c06dfe0a0df42374188680f11
204b7828d757326e1974612ea28c4d0efcedb7b0
115174 F20110216_AABTWJ sachs_b_Page_048.jp2
fb5aca950d5afc9345c8cafae4b1e813
b0f8bb9ba609f614bbd7beb38feb5a31914fa4f1
113210 F20110216_AABTVV sachs_b_Page_033.jp2
5841decfd126dcd9dccf70418167a573
dab141083ea47cea6e87d05bee0d1069eb7e47d1
2454 F20110216_AABUBQ sachs_b_Page_096thm.jpg
f70809e9681455348d825e9c8e6e15e2
6c90cb7f3aa28ac219e89a734ae5210155b0cdc3
29780 F20110216_AABTWK sachs_b_Page_050.jp2
bcfbcadafa8fea1f92f912725d34266e
ca89b6c1630ebb11f0a7f3d37baa05e0be82ced7
22080 F20110216_AABTVW sachs_b_Page_034.jp2
4c9a5b2ac9813528ad16a92bee7bb5f2
f85c4c31bf3b3a938a23ddb02137b58e6fa63d90
8467 F20110216_AABUBR sachs_b_Page_098thm.jpg
09a28370180414db7548a1cfc91eea9f
882cebe4dc35e6348de443d3614fc070fc78cc4e
103495 F20110216_AABTWL sachs_b_Page_051.jp2
8d93539f2f11b572cb757a458e12e441
1467a89a66da7297194522949b8428fb22bf773c
98310 F20110216_AABTVX sachs_b_Page_035.jp2
44aa3215aa477b3933f0034241094150
f000666acafbe1abd55068e181d910cb0b0fe1cf
122669 F20110216_AABTXA sachs_b_Page_068.jp2
4de15f5d63f525e699de8b16340ff1bd
2cdf158aecdd315a187dabacdb6368bd861b5d18
9281 F20110216_AABUBS sachs_b_Page_099thm.jpg
8f30670487e7dca702ad1b524e893442
6cdd281a1c682f8ecba7d0831373530a36c1a9aa
115099 F20110216_AABTWM sachs_b_Page_053.jp2
12cbad89cafaf6be694ac7a2abe153d5
f377102b4da46dc0c7b9fe9214181b722ad115e6
95834 F20110216_AABTVY sachs_b_Page_036.jp2
073d220e06ac42c94974d3cd101332c6
39a768ee8b15a5ea7803b88ae010256affa029f8
113732 F20110216_AABTXB sachs_b_Page_069.jp2
bb3ae5c8325cedeec2e40520cd9f8452
595c3d503450ac88d8f1b3fa5e7e4a418ec15abd
8879 F20110216_AABUBT sachs_b_Page_100thm.jpg
451e0cf582347429c604bf935ab4ea81
70430cd99d0bf70b43433ec2173ddcb2fb7d6ea9
115531 F20110216_AABTWN sachs_b_Page_054.jp2
be8ecfa9e08022b75e5c440b2abe7448
0ae0eab4847cf36d31f53ed49efb3f73f98da1da
111298 F20110216_AABTVZ sachs_b_Page_037.jp2
31fdc8c528f2a107f8c3a6292d0ff3df
f50aa259b453d6d6c6ab4ba378f8eb8ed1d32cd6
113523 F20110216_AABTXC sachs_b_Page_070.jp2
f646873a98389d64e277df41f87ca77e
25eecf2ae6fa484491899ab01b9495d0b33bcc7e
8822 F20110216_AABUBU sachs_b_Page_101thm.jpg
a06ce0ce0f9b57e8fac4f0da92342d9c
c8ce221359a5e535a8c95e1a81246f5bc4a651a5
116850 F20110216_AABTWO sachs_b_Page_055.jp2
511ba6898a9b7a39e4d3e7c51e005336
f4049e2d6b8d679e26b81f57a0c7204f3af24c7c
116567 F20110216_AABTXD sachs_b_Page_073.jp2
ed32b9d8bf73f027bda5e55eb60ab896
25938aa2b3379a7b53891c877c9f6fd6892bb322
9025 F20110216_AABUBV sachs_b_Page_102thm.jpg
333c5d57c72ff23ac9de5177c3289173
bbc0617f35cdc0269f74a1a95d273b1332397137
118389 F20110216_AABTWP sachs_b_Page_057.jp2
4d6684ed1d24854f2a7649eb70aba584
49f52395f112118f6cd864bfb71ff7c8bb2ad34e
118456 F20110216_AABTXE sachs_b_Page_074.jp2
6600389b6e842d48e9a31ffe0f5adbc9
6e11df594c81841e21f1164afbc19c18c6dcbc17
8921 F20110216_AABUBW sachs_b_Page_104thm.jpg
d607ba7c6ff76bd2b836a13e0f678c9f
426ccaaafb1fe00a2bf6db7316aff8589b7147e9
120810 F20110216_AABTXF sachs_b_Page_075.jp2
c54695ceedcb579883371acdcc3c37b5
50e36a5d16c4c19ecd3db4e21de64641cc497e67
9256 F20110216_AABUBX sachs_b_Page_105thm.jpg
56156b6342b8e5e60a61a87c073a39b9
bfeb8177f03036fec0ec1fd546d632b69dbd7e37
108402 F20110216_AABTWQ sachs_b_Page_058.jp2
8d64e0793db9bcd8d89955758a4c62b6
6557231bab925ae7aef9562222f5b03630f8db37
118496 F20110216_AABTXG sachs_b_Page_076.jp2
6779d7f79d88ce7cd2883067c4ebc4f0
54896a149ea39a00b565ae0fbbf80035f7984250
8782 F20110216_AABUBY sachs_b_Page_106thm.jpg
7db133c1c7f3643d0edd39ed01c1410c
067045c44f26e2bac929499477dbb17a706a732f
90511 F20110216_AABTWR sachs_b_Page_059.jp2
f08cabd4877eee5c12786a70631c68e9
a3a5ca4786671d71bd93ac400b60f26aa44e1dd0
118350 F20110216_AABTXH sachs_b_Page_077.jp2
ddc0ce2009fe0b9f8436f626b3fabfa1
bbaf958a43694921b6d96a33fc9c5d058d3894ad
8798 F20110216_AABUBZ sachs_b_Page_107thm.jpg
499c32c962f05e9b91634b736de05c4b
500f8187ddd6faa29bc5704d657321531666f5e0
76504 F20110216_AABTWS sachs_b_Page_060.jp2
1944b2c2d8f1555c142336f49c06c8d8
e6cbe921aebb3616be65264cef42c9db6d8269a5
117564 F20110216_AABTXI sachs_b_Page_078.jp2
b506370c56de02fd0f787911bb01a889
53d58429db83792490beaacd34465fe4970f1695
64880 F20110216_AABTWT sachs_b_Page_061.jp2
98553aa28445abc3bcfc4b6a69937a20
5392225845e95ff037f4eb0fa0ac922bcaf6de78
1053954 F20110216_AABTAA sachs_b_Page_106.tif
4cb5c37c45e242abc2aa951298c1e78b
544fe1205a328c9b867d8ed1a91fc76d98735f17
120227 F20110216_AABTXJ sachs_b_Page_080.jp2
2d834bc6442f77646517b9e633f6054e
e7096b26dd364ded7fb35c33330a68fa345a9e95
91926 F20110216_AABTWU sachs_b_Page_062.jp2
0f363414ab27aa660daee2786a118f93
63ba8d560a31c0c8f637e66ef042f279937b4772
119671 F20110216_AABTAB sachs_b_Page_021.jp2
bbebe1a09e9610fd1e74fed0be961b1e
01c71572a6a458ac94424a493280f17408fbcd88
121613 F20110216_AABTXK sachs_b_Page_081.jp2
a91aaa496317230c2a35e9b5644075c5
620aca1b2778d27ca7c8efc7e70716265cdf11a2
528548 F20110216_AABTWV sachs_b_Page_063.jp2
b1ebfb65477de8bef3788a4aebb25ac4
2aea9c52ee46789484fcf4e02e10e4113f502440
1893 F20110216_AABTAC sachs_b_Page_009.txt
22db1c240200d21a1a5227b4aba30e68
e0bd7e3f2ba662b7ff216a0c3b3e0907a083fb62
120731 F20110216_AABTXL sachs_b_Page_082.jp2
be7670f08fd674a4f25d16f732741f8c
4157d3e47e1d745df85ef6d9e4ae2a6f615c0dc7
180182 F20110216_AABTWW sachs_b_Page_064.jp2
e3fcfb4a4fd2fd06d1fd808e33f4e775
a5f84fdba053e3edec440c67f43078ed88750033
53533 F20110216_AABTAD sachs_b_Page_055.pro
33fa0f84aa44305a7bf3f4ff92ea024f
4246979aa93d0be50d0098ce16d07d5c58fe6e42
124902 F20110216_AABTYA sachs_b_Page_097.jp2
e49176ca0a60ed501d1623c2a4ca4b69
60bc71cc51e660389795556b25451492fc98e4e3
113002 F20110216_AABTXM sachs_b_Page_083.jp2
b7b844f82c0239c0cde1b0a20d95a065
08001a7bf9ec196a8905d13701f0953547b0f759
298436 F20110216_AABTWX sachs_b_Page_065.jp2
616f6ed2d64eac2b197da745f6734611
913d117b0bb4a8a2aee0b3e79a6434874e0ccf0b
104104 F20110216_AABTAE sachs_b_Page_052.jp2
9a66c6fe5d1a9e5a72aa9ecdfbd2d211
0a6e2d365ba584f7edf6ca06202c8addf367e9e9
116007 F20110216_AABTXN sachs_b_Page_084.jp2
eee4837a9516eb98a583c25a4c6b39c0
4078a11c5d5b796e454400c1b0a2fb3a208c1595
150384 F20110216_AABTWY sachs_b_Page_066.jp2
263efb4205b2539cb0ce1123f5dfa507
8af9b872475983e1f47e57619a5d72219fa538be
9252 F20110216_AABTAF sachs_b_Page_014thm.jpg
befd5e32970060a3b4fd1376da5a3f1d
b40b2a8a9ca5603c7f07e9f4931f70db33c77453
121774 F20110216_AABTYB sachs_b_Page_098.jp2
66b1b4a6a1ff6a2a3e1f281561ecd9a1
908faac4f1b855da051dd1eb3837267ce90acd1b
120470 F20110216_AABTXO sachs_b_Page_085.jp2
7d3b52783c91f44dc7ad01a44aebe3af
a8949371e9ccdab6682007adadb42b8de01d5594
110843 F20110216_AABTWZ sachs_b_Page_067.jp2
2486a5c039eedebd263fd1dd4be639ba
050631bd89d58057c4912e6c1f4c42d2efe14748
F20110216_AABTAG sachs_b_Page_038.tif
b9fcf3f60229f81fbaa97ac37d4c7423
d8d4b76f2dbddb19036fe3081f08769172d0f23f
139836 F20110216_AABTYC sachs_b_Page_099.jp2
0ef10bdc423fec69c3268af70e36b548
1c843216a76d3bd1fb8e43ddfab2edeb84780782
123252 F20110216_AABTXP sachs_b_Page_086.jp2
394b2d96eb0ad5a5de69ff27dbd03183
e18b8fa3c0daef7db7e1b3331d944a30216c6a0e
2077 F20110216_AABTAH sachs_b_Page_023.txt
8dad5cf27a8ce0dd296251d51b835f7c
c2b584d818d46d1bd80573ed2585e890548b7e40
125273 F20110216_AABTYD sachs_b_Page_100.jp2
4516b7abbe5f86c1817232627ddd4d10
d4ab1764d8158d4aa5fe7cc94838ea84dbd53dbe
119357 F20110216_AABTXQ sachs_b_Page_087.jp2
051867c2de5222cdc98a651f554e5b8d
5671dda6c783c23099fe2e50b588b9679897d54d
114552 F20110216_AABTAI sachs_b_Page_086.jpg
e82d3edcdd2cf1b6fbc520079882b742
033551b3dfc81369c524ecf6eb1d01000c419263
1051970 F20110216_AABTYE sachs_b_Page_101.jp2
44520c03af113c9124cca0d3eac006c2
a5fc91e8dc97365a24cfa62cb34acbf4ea868627
F20110216_AABTAJ sachs_b_Page_057.tif
6b7e19d530e38976d2dfbc63d8e40e63
e51bb7bf7dd79c44503b1a3295e663e8d63841c4
138434 F20110216_AABTYF sachs_b_Page_102.jp2
ebf9e29a98b423ed01b5945a795bf953
088113e243e353b94fdd80cbaca59365414455c2
120187 F20110216_AABTXR sachs_b_Page_088.jp2
752b61ced4fe6c82c589c0d508f93c76
b75dac794436dbc75756d14c3557cc8a501c4356
116768 F20110216_AABTAK sachs_b_Page_056.jp2
8d1331345d8fe84f358cddeb512557c2
13be28d2a1f7b9b86ac2b64dd9578fc088536922
128462 F20110216_AABTYG sachs_b_Page_103.jp2
9491d7c4af2a0af45813697998c52105
b1087465e53e4783321c90ee16e8ccd37a7b42a0
113885 F20110216_AABTXS sachs_b_Page_089.jp2
6e4d0f3165b74cef5ee47da084032b21
16a599820965f6941ceb8202aebd14a811e50e01
87247 F20110216_AABTAL sachs_b_Page_005.pro
2407ff637d23cf8c7b00adaa9c422808
ed8a862c1649795c1a38f86bb853de57024db45b
129226 F20110216_AABTYH sachs_b_Page_104.jp2
b555cdef15f0dbe4c724188a2d952c18
a34d27663f53590db3ab5017f615c42cb0fa815c
116127 F20110216_AABTXT sachs_b_Page_090.jp2
8c4758181e03d358a85d9c30d4c3cdd5
e2874804e0e2bd83562f2a08a9c4cf2cfae8070d
8548 F20110216_AABTBA sachs_b_Page_048thm.jpg
f202c619e6089999898d1e9d5627de0f
7191fffbfda7ff6f92759be187d3baf3e7149d74
56543 F20110216_AABTAM sachs_b_Page_018.pro
4231a1ed08b9c436d2d5409fc4ee79d3
d66ab58168702ee669542e913d2d9df1615ff025
137775 F20110216_AABTYI sachs_b_Page_105.jp2
0bf00ca1acd642430152c519c3717d10
af3dc91f78d6aeac891d519c123e35fe4763454a
114677 F20110216_AABTXU sachs_b_Page_091.jp2
343bfcfb15053c4904263c95849ee066
1cadea86e769ce4615a59b203c45766b6918471f
44544 F20110216_AABTBB sachs_b_Page_035.pro
9ff04642e13a10b7bd4016498eb986ce
70af211ff939bdf1e6f421f948a9481ae907fd65
2009 F20110216_AABTAN sachs_b_Page_042.txt
1bd8c3754e22c09e487b628a19b12191
df28cbd2ae47fa1deca56ea23edbdc7a58d4c161
135519 F20110216_AABTYJ sachs_b_Page_107.jp2
5ed87d694de082e56b82c1a7c65141f8
71a4eca64fd874d219dae22ba35e85934f8532ef
120718 F20110216_AABTXV sachs_b_Page_092.jp2
1012e8e7e62d50043778d3662e86b01f
01ebc8c8ab384b934986702a2d39de5d06584149
362 F20110216_AABTBC sachs_b_Page_064.txt
24f603c16301d5324ef8ce49ae6d0562
375392f59a3ace2a2ec5e0376644a0c45ef3c186
F20110216_AABTAO sachs_b_Page_053.tif
b050e530caa2df4a3295493ddca17b69
2b1035df5afd41bc6bf71d93d6674528e47dee6d
76090 F20110216_AABTYK sachs_b_Page_108.jp2
3c39fcd6089f26555a3b7b3313509b85
75923a57aee31e34147736ca797fd32bdc3367c2
48751 F20110216_AABTXW sachs_b_Page_093.jp2
025316b3c18cf86e41718c05a1c89c3f
259022a243ab76dcd85d3f1c33bd32c423382656
F20110216_AABTBD sachs_b_Page_077.tif
885de5efc7cb51df0de6785bea6a654d
6d56218b7fb4b7ef666d7531f7e7e16eb38a30b9
F20110216_AABTAP sachs_b_Page_080.tif
541a60a59b3dd6a1a31f3af825cbfacb
8a64a0d4b13a7099446fb650a2b1fe074b2e9b82
59717 F20110216_AABTYL sachs_b_Page_109.jp2
cc0f3763fd313c76f1aca1bee050310e
4e4be27d54da1c537ce2e807df3198bdd67a5710
138979 F20110216_AABTXX sachs_b_Page_094.jp2
670dd1b5f4f6666f499a61ba29795321
da50dea6832afcfe80212d4d5cc7904a4b13532b
57166 F20110216_AABTBE sachs_b_Page_081.pro
ed590311c1c7322df9f7e172d48d3fd6
6b5181b4c6c5c1a911c649c5b23286bbbaac01ea
2188 F20110216_AABTAQ sachs_b_Page_082.txt
f288dab95ba815878ba362608e7cc573
849c296e5f843fc770a77f526633a2ae53348173
8948 F20110216_AABTZA sachs_b_Page_019thm.jpg
62575c9f9a57814148a036713295d4b0
444333c2be26eb72721f3293d128e901b1000622
534 F20110216_AABTYM sachs_b_Page_002thm.jpg
dc2645358f95583d67bd9e0b5f186f2b
70edacec771309384efe27c6e747ec1f1732df2c
34355 F20110216_AABTXY sachs_b_Page_095.jp2
5bae8c3b744d80b006f5a497e1bb8ce6
48e4e160dcd33ad8a032412b152ff8e1d2e8817c
37659 F20110216_AABTBF sachs_b_Page_015.QC.jpg
1627bf57d7b04a729c5f6bd28bccfba3
f521669856b6765409d6a23a8e9962f48e250a16
30105 F20110216_AABTAR sachs_b_Page_001.jp2
7cbc08916aa59f0eec3db6cebc410522
0b076f5e2fe4d6ea5fafbec8cccca240879e4ea6
8587 F20110216_AABTZB sachs_b_Page_020thm.jpg
b291b943eb8a730be8a686bea9c7b72e
b6ef4f7f691f622773f4ff9202f53a4445413b01
1148 F20110216_AABTYN sachs_b_Page_003thm.jpg
120f8fcdb027a14f354237a2a57837b1
4b4f91ff52f718a779d3e9c6714eb93b5296826c
23085 F20110216_AABTXZ sachs_b_Page_096.jp2
7926f0e21c4c4aafd694f2703b9a95dd
0f9335ee3d2ea1ea8f6123c7cdd80e8b748c7386
2235 F20110216_AABTBG sachs_b_Page_088.txt
f9bf7d75fff8864355d31e03a2da84b5
f06b6f1d1898a3f4af854c911fc19645b79ff48c
F20110216_AABTAS sachs_b_Page_026.tif
396ec04cdcc25528de011f460a264d6d
02b73f63142a4dbfc8b8f065d7a96b7c6ad6a64d
8830 F20110216_AABTZC sachs_b_Page_022thm.jpg
153f744120f531216a4ce3b1e4c6928b
3b5003f8b225cd71c2723f04472851dae6a07afb
1834 F20110216_AABTYO sachs_b_Page_004thm.jpg
7b909019af5951181fec1577e8a46639
3e54179c9ca64cca2625056ce8d6e53280cfdaf5
8650 F20110216_AABTBH sachs_b_Page_028thm.jpg
384a15473b8a55ba40843d3a3f1674fd
ac9efa5b48546f9d132ba7035abc8b04f63f4111
56885 F20110216_AABTAT sachs_b_Page_075.pro
7e54eca9697f3997875cb4454a2489e6
0e0f91ef746f23096189b9873755c37ab46ab371
8111 F20110216_AABTZD sachs_b_Page_023thm.jpg
1144f440a34a71dcf5a97cdf1cb89936
690152bbb1437a3c19fdd3ff7a89b4560d5ac266
6116 F20110216_AABTYP sachs_b_Page_005thm.jpg
788f0d92cf9ec6d2e4ac721245d80742
def2202bcdabecd1f9d573cef1e9f754d0cd5a38
47526 F20110216_AABTBI sachs_b_Page_051.pro
4cd3bfb0b3ae7bdae8436e194a46478a
ec8c466e578e2cdea721d411ba7c026549772948
8249 F20110216_AABTAU sachs_b_Page_004.pro
78ed67891a648a7f6ffa2cc0c34e5eca
bd967c3c044804e6be74fe68ae2336f90cfc182e
9074 F20110216_AABTZE sachs_b_Page_024thm.jpg
ba226704d51ab50e380a4e92ce0fc91f
917c7cbcbda4ccfe7293cfcf47f49f792f458894
4235 F20110216_AABTYQ sachs_b_Page_006thm.jpg
f3f96ea542151c882aa1e0b43ea29506
8c8656fc87e4bd3f4fceb3cca5d20146bc103e4f
37914 F20110216_AABTBJ sachs_b_Page_081.QC.jpg
2b2c4a7f9a3fa90eac1039c0903835eb
b8876cadbfbd0789017b4818c1a9c73aa6cccf73
F20110216_AABTAV sachs_b_Page_042.tif
ce9f90c2f28ecba15caa45ebc9bda4d6
2d61425627e31935b46cb79ebe157716d1ccfe6d
8734 F20110216_AABTZF sachs_b_Page_026thm.jpg
3551e73c67ab5c90cea8990bc240f631
157162173276b2fb1b438be549d557c92b88099e
5028 F20110216_AABTYR sachs_b_Page_007thm.jpg
e69a4188c0b9fa70dbc5df74cc048368
56c1744c220dddb0eef97e1a42c01b52c21191c4
F20110216_AABTBK sachs_b_Page_013.tif
a8d786fba0e0a2418d3b67c607e00386
fe35702477e490b508477d0e0a127abb47adafcd
1986 F20110216_AABTAW sachs_b_Page_058.txt
f7470e1b2684eb9427525a50746b17b6
a74490150ca9aa3272dfd8a1af687220e557d65c
9138 F20110216_AABTZG sachs_b_Page_027thm.jpg
7d6eca513639a464386768d554679d09
7d794bb49b7584c413762ea2326dbcadce8451c7
9050 F20110216_AABTBL sachs_b_Page_103thm.jpg
e9fb2dcee361df435bf7c2e184c73522
912329b6f5b8a036d725818732947d3358af5517
2160 F20110216_AABTAX sachs_b_Page_090.txt
5a75ef2533a9210dcb440b5492f51ed3
212411c2899dca07db5a80754b5ade552117d242
9089 F20110216_AABTZH sachs_b_Page_029thm.jpg
feeb63b4ab88e2c7330255b9b84a2c11
8bac692888bdf02a53640612e779aa72bc34337b
2492 F20110216_AABTYS sachs_b_Page_008thm.jpg
d13b974485be70f88659bd380d1d3c86
31fa90fa7531a7b0ec724bf2bfbce544d0a1ddb5
7703 F20110216_AABTBM sachs_b_Page_064.QC.jpg
f0c692fdda6f97a71fc6e07120303032
1ccac625b6a73eaaa83a9f633cf94ed0f49395d9
55595 F20110216_AABTAY sachs_b_Page_013.pro
909cadf1fc08001c97d59497ffbd391a
42293345e274e0a54e0f2b11b1f970d232dd1243
8990 F20110216_AABTZI sachs_b_Page_030thm.jpg
c1e1ef56f1251b99c661846327c08350
7043a3358054b74363178b21072bc15027e9b936
7276 F20110216_AABTYT sachs_b_Page_009thm.jpg
d3259b8225b23fad8360808a464b2e14
865196a762a41dedfd0772fa12ada67827ab1a8c
9014 F20110216_AABTCA sachs_b_Page_087thm.jpg
3693a860717cc86d894f98a6c0300212
761a808afe6bfe92d88948f559efc3712f1ccf41
9177 F20110216_AABTBN sachs_b_Page_018thm.jpg
361b62a5808ec865923c940e5c15fbc5
c3319d1ea51b57ad2672678d56f48f26f7215e52
115209 F20110216_AABTAZ sachs_b_Page_072.jp2
4270fc8f45b5aa008538442c421d4a3b
fd34bd2664ebadd2a4c75f75ace6c0631d081226
8862 F20110216_AABTZJ sachs_b_Page_031thm.jpg
082b8abf04462d898ceadf483bffb277
8954d72663213cabba30b3bcf5d60f2c9ec4fae6
5456 F20110216_AABTYU sachs_b_Page_010thm.jpg
a017c19371b6573338baae65a8adac08
6ba0855ab59607ed8335fa433d560596d2c1aaf0
8873 F20110216_AABTCB sachs_b_Page_078thm.jpg
d3a3105db730b8321bd5d93c2928eb10
dfa7ddf51eb6d1bfe7e2ff42a007bf3815d05102
F20110216_AABTBO sachs_b_Page_046.tif
3d33489fef6dcad813530267ee098a30
a500eed9bc6d779cc4fb11fbdcb55ade93a3beda
8827 F20110216_AABTZK sachs_b_Page_032thm.jpg
898b6b4ea5a14fc8d9d980b2a1e3ff86
93e9f141190cc447a597c436516ab0d7b179e6f5
8907 F20110216_AABTYV sachs_b_Page_012thm.jpg
707aa9ec5ab72d3894420ef09ade8eb8
293f0d2fc7c9b53f3704b570a2dab72e8f7c442f
2570 F20110216_AABTCC sachs_b_Page_001thm.jpg
231f8a4b4370b5ac62780aa86267282d
d58c865ca6e1d65153ae70b9855533ab0fa010e7
121401 F20110216_AABTBP sachs_b_Page_079.jp2
13695c8834e5fbe8e24e08a2244feb8e
2c5502a4c2adfdd906ce1f336fcd96bd38bc0e0c
8591 F20110216_AABTZL sachs_b_Page_033thm.jpg
e1e4f70a938a19b5029e8a3111a43ed3
e434d0cf282d60f066fda7f06c51532b0f2441bc
8973 F20110216_AABTYW sachs_b_Page_013thm.jpg
49c78b7fefe18ed0d5b93c6cd513b492
4a69fe805c18a8bf77c1ad44bb9c54db7c2f5127
130942 F20110216_AABTCD sachs_b_Page_101.jpg
66f5a56ced48a342e83e7e6cbd7b8803
5eedc92bab555acd1dbc433a43e5b084078631b9
F20110216_AABTBQ sachs_b_Page_014.tif
0c004b3d34d3652558c69d5e956e82d8
bfde45871c70f4cf8c2b6260a451458e253d4445
1895 F20110216_AABTZM sachs_b_Page_034thm.jpg
9783dbae97ee621bed15438dc6fa006a
8d44e6353590f0801f1cf43654f6cc8c5823c395
9150 F20110216_AABTYX sachs_b_Page_015thm.jpg
bcacac8287609a4e4ff8d19fbf1b1966
db392616725ba54adf3f705201708806f8cc6c78
54757 F20110216_AABTCE sachs_b_Page_020.pro
a707025a31e7b2d879eb39f2dcef858f
a53817974919fbf8c1326d92da8f37651ea734b4
8909 F20110216_AABTBR sachs_b_Page_025thm.jpg
8a037e72c698a3e19ec268bd459e8101
dc8dc60581458057797330a0142a920e1aa026c7
7222 F20110216_AABTZN sachs_b_Page_035thm.jpg
dc1face38fce0a970cc867e5de6d1280
5267255de1c7a1a162ff991c8fa7585e27f2497d
9000 F20110216_AABTYY sachs_b_Page_016thm.jpg
808e0e6fbe279ff7af0aa60e4103975e
11ee77e120749ad8e9980117c9803368a58ccae0
176048 F20110216_AABTCF UFE0024780_00001.xml
a1a47f6adfd2303058531667fc727e1a
29af275fe59f55deed6658037c04fe0383cd28e3
119685 F20110216_AABTBS sachs_b_Page_016.jp2
ac71eba99c3d94cf6f03318353d07721
ad4e5bf08aebcf0388a85b04706b5895bd466b85
7008 F20110216_AABTZO sachs_b_Page_036thm.jpg
dca2d03fc135c268347b66156cb88ddd
85fc19aa67941157ceb48f11dee47687e8468fea
8490 F20110216_AABTYZ sachs_b_Page_017thm.jpg
08e3b0583f010675646a3a13ca47be0e
491a9e2ab66065552fdc3041b32da9422d75518e
2046 F20110216_AABTBT sachs_b_Page_041.txt
2e2bbc99d081181bf49da87adf7bc5de
1ab072d3efdca103d7ad9db297cf7cbed082eb24
8509 F20110216_AABTZP sachs_b_Page_037thm.jpg
3aeed78856c94c7d26a686e11070041c
6fbf6f9812c3211cde101a74845a27211baa7aa4
F20110216_AABTBU sachs_b_Page_056.tif
cbc6ef436be4ba9740839e4684fcec98
a671024950beff26eca1ffd9553502bc62784868
9136 F20110216_AABTZQ sachs_b_Page_038thm.jpg
059ec65c11b89b1b9f12c2bc4b4fca1d
8f376e6cf4a3e7f9af08334151fb83e37f4e00ae
F20110216_AABTCI sachs_b_Page_001.tif
e4438ca8b3f1965f9a5aa712a3e95b06
89eb94e65783c2b033130bb5d1b14d4e9027f483
109164 F20110216_AABTBV sachs_b_Page_042.jp2
4ce096daae9ce605e6b25adde850878c
3d9a02d31b6ca2e136aeb36832b703b82aefcb54
9060 F20110216_AABTZR sachs_b_Page_039thm.jpg
4e72a87aaeee18ed59ad56a30df35ea6
81f59e303482bb13de6ba1a41e0c140e6feb34af
F20110216_AABTCJ sachs_b_Page_002.tif
56691b5e29821dd47f62fb55809d15c8
af04400bb66cfaafe87a9333cedd14add96962a4
F20110216_AABTBW sachs_b_Page_032.tif
b7a430d8b2a761226315770b8d67fe9f
ec325b124d81ad0af0229850d916b03ecc945dca
8894 F20110216_AABTZS sachs_b_Page_040thm.jpg
bd912fe5ae9aee81fd4bb49dafef8665
fed5dd16e1fd4b483510d31b5167469256a42951
F20110216_AABTCK sachs_b_Page_003.tif
10914b11d958b3612d3c8e16c008a33e
8a2841f829db1b68df000e72deaf314d41425707
103926 F20110216_AABTBX sachs_b_Page_041.jpg
07dfa5464618be47eb3f496d00274f45
b1c06cdd9768e480848fc7e41a8bef7b5aca03f0
F20110216_AABTCL sachs_b_Page_004.tif
07d3b271a452751c8189edf0c4899c39
4a5c360b4c6b7a91fa6138794a6805b374a20af8
128605 F20110216_AABTBY sachs_b_Page_105.jpg
6240ef9705f66f95fd7d0d18d16860e4
d2e8d2a070d46992696ba0f12e6b5272261f5526
8811 F20110216_AABTZT sachs_b_Page_041thm.jpg
d31d8a9b22a0fc90a94b1e38a6d7f771
72df2a1c3d854b25fc0b5193085ef68fb567d942
F20110216_AABTDA sachs_b_Page_022.tif
8c53d5a5927ea78d643bf4d9e2cf2bf9
676ea5e9e0d765db24e8f688b4d6cace058d7a0d
25271604 F20110216_AABTCM sachs_b_Page_005.tif
071a855c23bfd70e32b603eb304205b2
a612ef327f7a6bfc401aea80484b7cb08d93682e
F20110216_AABTBZ sachs_b_Page_089.tif
de4cd320fc48e5dc23772f4088b08334
127b371ecd0ee166902f3df0fb5db700dcc1f5b3
8574 F20110216_AABTZU sachs_b_Page_043thm.jpg
279e1f8d5466d3bd52dda3880b0c3b2c
7171b9ad218be53c0cdd8aeb05342280ac0da9a6
F20110216_AABTDB sachs_b_Page_023.tif
c0587aa1e8df869ec68b257faec15601
f635e900c4f5fc0369cf63e42d20e9e7d909815a
F20110216_AABTCN sachs_b_Page_006.tif
75569f0bcf3fceb3d8b64a8a6e6eb499
7c329b005afe26c5d0828680d7d307b302dc86b1
8989 F20110216_AABTZV sachs_b_Page_045thm.jpg
d5d4305ab569311f7d17fad19e1f1bdc
832d63840e60ceebff9bca9384d6ddd35e1b80c4
F20110216_AABTDC sachs_b_Page_024.tif
efe7ad5af8759519ad016f4fbac3a748
5df6f8b1345b84544fed0cc3a36ba0c2daad9aad
F20110216_AABTCO sachs_b_Page_007.tif
006cc2c80363ee795d018e566aadc1fd
a7d60b2588f474af89dbb1cb700d01939d6bfd18
8628 F20110216_AABTZW sachs_b_Page_046thm.jpg
15e961572bcae8199a43da2c3887bc23
d3c20e73bcfa2a4744566f3ff3007dc5db828f77
F20110216_AABTDD sachs_b_Page_025.tif
561e390c83ce5ec7346ccef8b90cecd4
12561e6cc17105c6bb0c988ebfc2b8928e2bbca2
F20110216_AABTCP sachs_b_Page_008.tif
aab480c984bafb74f1f82fb360fca4d0
8e015ce5bf66170cb9bdad16a4f2fc66bd545fd6
8849 F20110216_AABTZX sachs_b_Page_047thm.jpg
d924636fb97b8815b1644d2af430023d
df7111602c91045baa45909aca0065bb7185a5f1
F20110216_AABTDE sachs_b_Page_027.tif
898d00e536d13298f1048bc0950e896b
a9829478c0af6255e761e42297e23b5e98f806b4
F20110216_AABTCQ sachs_b_Page_009.tif
ca44cb08fd2d03fd104c752f50545492
3412516a3297ecdf014e0ff654b8eddce2f6f104
3262 F20110216_AABTZY sachs_b_Page_049thm.jpg
3f076b01ff1838e90591e63478f1a70f
d600c54a4bc76d9b17ba02d3bb7f2758057e34cd
F20110216_AABTDF sachs_b_Page_028.tif
686d579f24406da6e300477540a14668
c358be1107f908442ed6da5f8034f8e638f5a708
F20110216_AABTCR sachs_b_Page_010.tif
1147411544c387c031876015a45b5a39
7824263895755b61ebe98861952fdd8f06fcb4d3
2472 F20110216_AABTZZ sachs_b_Page_050thm.jpg
35fec92e3885bcabd5c0111f15455768
a1c58288888a853609d2c0584667dc93d5ca5e50
F20110216_AABTDG sachs_b_Page_029.tif
a951918ca38cdaf4fa23aa655bf16f3f
509cb215fe1a2f29d0c8a4e4671edbd9140dc4f7
F20110216_AABTCS sachs_b_Page_012.tif
47a5a7e18fb8fb852c60f87a170325ba
d39fd20445cad2764da47b8ea7c3adf32df4ec21
F20110216_AABTDH sachs_b_Page_030.tif
401ba05f9d392fd2b4d85dfa691fff8d
42586aacea0e5613159d893170052b4e0d299ca0
F20110216_AABTCT sachs_b_Page_015.tif
dc048d6755740bb778555c4e8b23a9fe
7ae7099bd04d9b14268666a17760f60d0c14a312
F20110216_AABTDI sachs_b_Page_031.tif
58c47bfdb704aded07610affd884c27e
81283a0c4a20dc4d41fd563980ab3b8370ee34cc
F20110216_AABTCU sachs_b_Page_016.tif
0b36ef55334e96804e8783d08497f8a4
7fe0e3278f3d529aa2147a965f66af4560c044c5
F20110216_AABTDJ sachs_b_Page_033.tif
bd74815163c820b74dbeb77eeb186194
9b5c4995f7015f4d6e31a9e0439cbc936091317b
F20110216_AABTCV sachs_b_Page_017.tif
8ea634c5cc7a7e90ce1782474cbfb374
a39ccf34cb0f02e1b8f3ac21cef197c7ee3e6ffc
F20110216_AABTDK sachs_b_Page_034.tif
db7fbe317e3ef028f42bab6c961b0eaf
a7b41294661a4fc0e94a07af463f9e8c752fae9a
F20110216_AABTCW sachs_b_Page_018.tif
8abdb8f5f1be62bb2ba76349079979bf
b02966d4613d35ba48000903de025b55baf2eb42
F20110216_AABTDL sachs_b_Page_035.tif
cf5e8e2df30b901acdacdde294da7f62
7b9155e1f8e1c5a00dc545093c1f3552ce5a092a
F20110216_AABTCX sachs_b_Page_019.tif
e1e3490a9330e93198aa1c9712db0df3
3a67194a0a2c7423d7abfe79a20d6883f89c1061
F20110216_AABTEA sachs_b_Page_061.tif
206483c350bba1314597a57880b22fbf
d8d3edb916999d05c66e21f5b4d490dc9d210901
F20110216_AABTDM sachs_b_Page_037.tif
bd9ade590608a1432ebcc48bfa25f27a
9188016be0903e55fbee5ed9edfbb8b792076e89
F20110216_AABTCY sachs_b_Page_020.tif
6f8246ab45a2d5247a7cf6e3f1195dc7
e1b9739fece1aa7a42eb7b478f636e62f3467b35
F20110216_AABTEB sachs_b_Page_062.tif
219373c1d929ce44fe1e731c15ce48d1
c81552e9daf0f767d61d2494f76e5ab0e9025a3f
F20110216_AABTDN sachs_b_Page_039.tif
6bd971e733c00bc2d9f913ef429dac69
ebbf4c829efc7a6dd42650786587ee3570584ba2
F20110216_AABTCZ sachs_b_Page_021.tif
b147afab4c85345c2a6cf386b15618bd
1ce3269c99f3ba48d77afb5b439d35892882e2f5
F20110216_AABTEC sachs_b_Page_063.tif
c40b1010f28521333893c354908f2179
4c81b493542492a3d67da75a2bc26e259bc37a9f
F20110216_AABTDO sachs_b_Page_040.tif
6400845a284114a5785f485500c46f6b
b2fbd1fe4b722e9f88df7ea7c1cfa91fbae21860
F20110216_AABTED sachs_b_Page_064.tif
665fe8507aa9e52d5897c88a58366226
c02f1727d3da6a2682350d5768d7034d0f9da394
F20110216_AABTDP sachs_b_Page_041.tif
707c25fc094625b8e6e435188f1d4a27
f56fdf94ddd221a373ef0db60eec345489396474
F20110216_AABTEE sachs_b_Page_065.tif
fecb32e02d9a23db8687f37e3aa67d51
41f76fe7529c8b84a8314280e34d54c5084ddf15
F20110216_AABTDQ sachs_b_Page_043.tif
9a45ee45c963df48419ea8e4fbc32e08
2ce6fbea9d3888df8093748a4e92b055e6e55967
F20110216_AABTEF sachs_b_Page_066.tif
ede5637dbbf03faf02b26e2991feb5ef
f5c46dd55c4e51e9fa2754e89695cf371c4684b4
F20110216_AABTDR sachs_b_Page_044.tif
782b64b4ccf826ba8e9299ad3153b5af
a9043dfd473c89426e572104de3061a8a95b27fe
F20110216_AABTEG sachs_b_Page_067.tif
c325dd81012a8d6844bec4482a73130c
1c0facf7f2c3a0c33437a3a99b3fb642923de54d
F20110216_AABSZA sachs_b_Page_011.tif
a9737b4cbed604c555bc1c83d48eed4d
175e80e2ebdeae244ce7fda81e01e3233aa1e442
F20110216_AABTDS sachs_b_Page_045.tif
446dd4901475724df26b82634790aaaa
cfc68985583b07ffe68bbfc51be8cb3eb31fb717
F20110216_AABTEH sachs_b_Page_068.tif
38f8d4fd60c4c1681d9755c37fcb9cd7
84291cd4d9b6259a7d3b0937e69c56ecd573f66e
52278 F20110216_AABSZB sachs_b_Page_023.pro
faa7b0b59aef21f695ed50ac9bdef2e6
befcc19d7274396c056d4736899606366e145972
F20110216_AABTDT sachs_b_Page_047.tif
234748fe4f5c2c862ae6c0b33565a302
444541591b4132cf98e3190f036d643bdb4326aa
F20110216_AABTEI sachs_b_Page_069.tif
7597208ca0bc05e52af3f1b4550cc507
824f562095a242ec4746e3a02f279a861239ceaa
106477 F20110216_AABSZC sachs_b_Page_011.jpg
f7e494e503ec366bedf3f74d3da961d5
aef5a19b4914917c163f86efc0f2508e90c6adef
F20110216_AABTDU sachs_b_Page_048.tif
c1ef38507f10e623d401fa8a97c710db
873e08bbe820e7006abb67c776e3704f0f67b377
F20110216_AABTEJ sachs_b_Page_070.tif
0c2233c548d2a509b356892d6a42f10c
ac8bce0a80219a7f09698f9acabfe187fb5eb878
2080 F20110216_AABSZD sachs_b_Page_028.txt
36f580b60eb6f49c4f19a563f02b8dc3
a2d08806c7a92073a98cb483526184f1235a393a
F20110216_AABTDV sachs_b_Page_049.tif
a8332fb637ad26a35e9e664f2e22c562
0c6e848fb1fbf9adbbf2c2116ae7bf8b6782e681
37187 F20110216_AABSYP sachs_b_Page_082.QC.jpg
a2b0f2f49c5acba5ef6f884dbcbd4881
dfbe95d79891dbfb5b997e8764744a947cc7ee4a
F20110216_AABTEK sachs_b_Page_071.tif
97be107b8b6fb23a9f39b8a4e2faf4b9
3bec08f13db9c2a5ded330fa45c1aa7201618a79
28157 F20110216_AABSZE sachs_b_Page_009.QC.jpg
dd0939b4384d4f4a2b0c8bdac3b73254
a6254f798203aef00e3b67d37bf706c8726723ca
F20110216_AABTDW sachs_b_Page_050.tif
f52ad1ca14ed2f1863db72b91843f868
c61169e7ab896a150bf32cb590f93fed3d7ba3e2
102758 F20110216_AABSYQ sachs_b_Page_043.jpg
bc60e95306a456aba0f27165145106f3
24cc5ae564939f4758de4fd899f6b6ef08a03205
F20110216_AABTEL sachs_b_Page_072.tif
c2e03993f2e4c4476529331fba2e0078
594c332d2e5a6399fba0cb6e1efe4a80b172f850
36072 F20110216_AABSZF sachs_b_Page_076.QC.jpg
310a96b84502361e637939fdf43aa53e
36371aa599c941989e7fd52256c6b02d9b298706
F20110216_AABTDX sachs_b_Page_054.tif
44a86082d215324bebc54dff68527e38
b017156f59eb46dfe22f346eab6062c103efbeff
2473 F20110216_AABSYR sachs_b_Page_103.txt
5c2d846b8efe4bdb47d7fdcaa0a96a19
89d3a02dff01cde1aa1277e69c1df838d2487c83
52228 F20110216_AABSZG sachs_b_Page_017.pro
cee21282c0febbacec0d53ca84fc05a9
91adc0c102a8f62cb6352078e8baad5005877857
F20110216_AABTDY sachs_b_Page_055.tif
8a8b080f750b921f2d359c6bf2d5b58a
3de4ed77166249509ff34f45429b1b4a21fa9407
35813 F20110216_AABSYS sachs_b_Page_073.QC.jpg
00e5f858e5ca91e6c673b27b0163ffb1
3d7792958e66ddc55f7914f62f407bb280e6db01
F20110216_AABTFA sachs_b_Page_090.tif
919264243ac549653ba3087eea99faa0
f61c8129aee995676c84d73f2e44a5e5014c5d38
F20110216_AABTEM sachs_b_Page_073.tif
72b2609af7cc67fc209c29187095846a
a39504a4d1271997a5ff51e7df2293e3e00225a5
F20110216_AABSZH sachs_b_Page_051.tif
64df6245a24e5b6e7188fdac203a3c32
c9f84e48bf19c2e06b380f6a4184800c3f76aa0c
F20110216_AABTDZ sachs_b_Page_059.tif
a35cbfe3555b6c7583d5aa070e587729
5fde7f16fd13c520e1a7358d255325075029ee64
F20110216_AABTFB sachs_b_Page_091.tif
516b75d6f47123c08c2256a83d5bfdb8
4c502e7ec7f75bee445fbf5c2f37c7ca5110bd25
F20110216_AABTEN sachs_b_Page_074.tif
e0a5b252c86a3b1a8fc0f74bbffff1d1
7c89f3f2b772a7035b51abe53b7bf505e708acae
1219 F20110216_AABSZI sachs_b_Page_002.QC.jpg
25e0510b0a23528887294a17b2375ed0
a5ae7f872b04009aead46cd08122b9d36ea82800
117040 F20110216_AABSYT sachs_b_Page_071.jp2
6b03df23e5430962ec043ed7161c1de2
4e63efac30aa343928fa2273f09ebdc21ac9c045
F20110216_AABTFC sachs_b_Page_092.tif
cdc9c91fb96125a5b4818ccdcbc7c19c
4bc076d48e4dbfd7b86f9ab0e88cc333c922a996
F20110216_AABTEO sachs_b_Page_075.tif
078a17f379c03f8ea4553d09e21bffd1
88bf325819fdcb50b359d6ac88311802d094d4ee
128013 F20110216_AABSZJ sachs_b_Page_106.jp2
0a17f26421db0c72695345f0735e88c6
bbd13cdd40e3359391785c29fca9be9a5d1668b8
9161 F20110216_AABSYU sachs_b_Page_044thm.jpg
00b3ce0753f402e0c630462f56b9d40d
850d954b9637c56dad1aee3615b426afe2762c9f
F20110216_AABTFD sachs_b_Page_093.tif
66253b7cf1c087a87d2f01cee371f609
28d1bdf3672a506025886b9ba5212554684119f0
F20110216_AABTEP sachs_b_Page_076.tif
f90e49ccdfcc791657c979ecadc6a22d
f90a8329094d9848f97cb4c9aeff73b018801dde
138179 F20110216_AABSZK sachs_b_Page_094.jpg
395433d2178642bba382b1bc6bb91489
27e7afc26b2215c210c10ea7cbe997cc4a0f9495
33784 F20110216_AABSYV sachs_b_Page_052.QC.jpg
61eee26ae503adbe28ff5b93aef6ddad
c9a0132f1f87938c4e4539e9eca126574c5437a3
F20110216_AABTFE sachs_b_Page_094.tif
7c692182df05b35f8308d3bd7865fd68
9096c5a03454a15607a64dca12cebe54ab5e60cf
F20110216_AABTEQ sachs_b_Page_078.tif
c85c642bc65fd878b2c2e5fa56701710
0361f42300210c107d448aff912f411598e9e94d
F20110216_AABSZL sachs_b_Page_036.tif
39cf234dc407feed323446460eacaad7
96ce3a5d8020bd275b838902777df40d01d4a29a
8555 F20110216_AABSYW sachs_b_Page_053thm.jpg
f23191955a65a5653bde3ceb98e33765
b3362c529048a93287fd5527f1ed0b35442f869b
F20110216_AABTFF sachs_b_Page_095.tif
1f7f5314287688201ba591b59c11c956
4f8148dd245fd3fa3d8b05557d8ebb144a9958d6
F20110216_AABTER sachs_b_Page_079.tif
54be55a36c539c7dcb899ddbea7f32e7
0b950dc3eb1f08e6541b9d1fe4865c57c05941fc
8781 F20110216_AABSZM sachs_b_Page_011thm.jpg
be329ed7cc5713747de7b4fbb0a4800d
0140a588a0fe3886399ebf9293e0427d69de8efd
1255 F20110216_AABSYX sachs_b_Page_010.txt
cc6eb92456e03114c8b783df844f913a
47ebb83e68e75a42f1daf10377d11568e606adab
F20110216_AABTFG sachs_b_Page_096.tif
d9f34fed64b69aec04c8d122b6e5a419
701c536aedaadc93322d4e77eda9d523f59cf684
F20110216_AABTES sachs_b_Page_081.tif
ac63691fe4393c5dddbea398ad2026e9
39b484f582f666c7e1f3d5676597b7ab6221612a
F20110216_AABSZN sachs_b_Page_060.tif
e4eb65a368260fc132d5df456a05e867
8c85e79417b34ec0e9ea122e7b9dae48b62d66e5
F20110216_AABSYY sachs_b_Page_100.tif
7c09125276b0fbeb8d1f48ac1e9635a6
f5dec9618ffe1843604491885657f9afb3300bb1
F20110216_AABTFH sachs_b_Page_097.tif
48a377080cb2bbae3fe064eb39b75ba7
f2b810972c4ec25849aaa55138f934216581832d
F20110216_AABTET sachs_b_Page_082.tif
6d92d01bee577d862068531891c6db9b
d8adac1801af980744bb0e8d08b58c22673713a8
F20110216_AABSZO sachs_b_Page_052.tif
e181fd4db8b8b18f772c660fbe291265
bdeebd6bf5ef07ebab46721fd7e88b37386d0fa1
57246 F20110216_AABSYZ sachs_b_Page_092.pro
ea12d9efc8bc1a98f9700e4e46bbc294
80fd8b98072033f9f53fe7528f153eea99c208ef
F20110216_AABTFI sachs_b_Page_098.tif
601157f8f77995b4167ea2ba9b6a9741
b67dd4888029e7f29b3971948a1c62c50d2aefca
F20110216_AABTEU sachs_b_Page_083.tif
76890d7426cc67fd7e565ec36cbf50bf
4aa1dcb1a1127be6ff25effa8f97904d76c3fe2d
51057 F20110216_AABSZP sachs_b_Page_006.pro
94a64f6f3853b0affb88459f1b0be508
579c94566bea7326f572bf1df0cfdcc1cc361241
F20110216_AABTFJ sachs_b_Page_099.tif
33399aa7d447b1858068e64389a5ca10
aafd757e565439a2b474b61c01af066b9508f9d4
F20110216_AABTEV sachs_b_Page_084.tif
1e399c223c67c813d9393d7da209c550
b47df2efb8a394a3fa18f8348e0b81d9159b8119
2223 F20110216_AABSZQ sachs_b_Page_015.txt
65ae731f56b800eb9c44efe643fb7b03
47502a7942b551227e7833c3ce6b0bc80199cdf3
F20110216_AABTFK sachs_b_Page_101.tif
4ea9fb91df74dd639caa1f3771bde6f1
b5e668dc1827634e00222937538909884d6fe52c
F20110216_AABTEW sachs_b_Page_085.tif
f8a74f12de7b3f76c2950913f59fca5e
903d5b43263e953f3ac10c3e0c56064ad850f0be
448209 F20110216_AABSZR sachs_b.pdf
53c5c98b92c01c38c0beb6b8b3866267
a07d2d06a3330b48510ac3963d052bfcb7475ee1
F20110216_AABTFL sachs_b_Page_102.tif
cbb9af6f5dfff0a70c530c7839dd0dc0
0128d8f8a6d83f6cc5b08bda40911d84f19fc606
F20110216_AABTEX sachs_b_Page_086.tif
8133f8dceb2d37373825cbfe0e3c89ef
a891d717f21d59200c977b1d669e86deaf6ed457
53897 F20110216_AABSZS sachs_b_Page_073.pro
531f9aaa845e22b6990e4aefea5a4f27
68f3e3e45d03f688e5001581cbbb00cf44316d45
2135 F20110216_AABTGA sachs_b_Page_011.txt
6746e98282506e33ba3ad4ab72ee856a
6c2ad833c8aba0e98a048031dbf7bc10a36af007
F20110216_AABTFM sachs_b_Page_103.tif
00a3572d40dde12a9cc185d61cccb738
d01033a3dbfa47ff778dd90ecaf50e1e3e02855d
F20110216_AABTEY sachs_b_Page_087.tif
0890b0d0ebe001d05fc66cb684d85fd5
ad479b7c0c885db184eb65f0c42f8073ead7ff8e
8177 F20110216_AABSZT sachs_b_Page_042thm.jpg
936be4883cf6b8efdcd5482afd55afd8
8d01f535d9787305cb09b608ede053b149bca516
2129 F20110216_AABTGB sachs_b_Page_012.txt
3858511a1d10f50ced24787edeaffd04
95b7415d60dae92e502fba6366251e9a2ec03749
F20110216_AABTFN sachs_b_Page_104.tif
909cc34a1e4bee8e2b39a3df2e0f656a
98580e5a9937297d7b14f72b19234b03ab3816a1
F20110216_AABTEZ sachs_b_Page_088.tif
011a471793fda7a8b77b8ce996f21f8a
d9d0f89bb6cc6b304917be1a6630e6dec49d95a8
2178 F20110216_AABTGC sachs_b_Page_013.txt
828816838b7903c63597a75142b2d0cc
4a096705ab3952d7b185d852cb95b95a2d049239
F20110216_AABTFO sachs_b_Page_105.tif
731c1fef11d96bea5ff947bea1e65e97
62ef6d7b38da07ba87f34e59e77e9e3db7f8f874
42743 F20110216_AABSZU sachs_b_Page_049.jp2
581b17ddbc15e44eb84785db4d7f4a18
3c7e6d2f07848ebd4a18edfbc999f0f46cf843a5
2323 F20110216_AABTGD sachs_b_Page_014.txt
15d8c947114163d81fb806a689a7d655
8c43c846aff22d9d78136ec37880ef4557b32ae9
F20110216_AABTFP sachs_b_Page_107.tif
12e378a62dd63ee099822fc92fdb8949
0d802f498255e95ae3e6017b1b6aee890c8d31b6
2634 F20110216_AABSZV sachs_b_Page_107.txt
06819656f687a14c16b595ced8d804d2
0db78b95fbc6906a514243dee9c93926e086ae93
2196 F20110216_AABTGE sachs_b_Page_016.txt
e2085b4a1adebfaf5b2ba2f35f0bd669
a47c7ac7e5976f7368e1f82c035d868df3ec8c28
F20110216_AABTFQ sachs_b_Page_108.tif
ed63f7456ef365ee6e657badd01d90c6
45e5608e327cdc91dc9dcfee3d67577ca930a203
F20110216_AABSZW sachs_b_Page_058.tif
55acdc7cc4815e3be752558e4503b4c4
8bca10a8226b85f46e2011d7e5eb0c91c6677453
2054 F20110216_AABTGF sachs_b_Page_017.txt
b266b64b99b759f3ed14fa919cd348c2
1adc08d34493fb79720f367422589853866e093e
F20110216_AABTFR sachs_b_Page_109.tif
25645b37855f5e2f0a8e26779c8da661
6ea19d1ed7c3b7b1a250df62b23b67f2083fc05e
8054 F20110216_AABSZX sachs_b_Page_097thm.jpg
8bab2b3e8e52a92e31458e55ca1267a1
0fc3240a1a5bc1d1984da66b31b56622dba3f43e
2215 F20110216_AABTGG sachs_b_Page_018.txt
6284693a836e682584e4dde848cf1bb7
73be90466a3dd9444007bda61349dac57b6a63e8
570 F20110216_AABTFS sachs_b_Page_001.txt
a6bdd64f1dfe7f20b2d388f83d201260
ae23be5b5a720902046561fae615d9dc3c6755f4
40191 F20110216_AABSZY sachs_b_Page_059.pro
61431d57997035cffb1b2ea82a5c3ba1
f7385e0f979bb2c26b40a86df38cc1e51988e0ba
2131 F20110216_AABTGH sachs_b_Page_019.txt
aa32772050b44eea563f4aee83ac37ff
8e6fd2fbe8f27a0aedb1e4034412f961a0c60109
98 F20110216_AABTFT sachs_b_Page_002.txt
0098fde1257d68cd95ddc41931ad88d3
a04a4d1a091e531c7643eea11c6a73a872cda34b
8838 F20110216_AABSZZ sachs_b_Page_021thm.jpg
7ee70417311f4c5c8c259b776aea2f9b
61eb39ecc9e15d378221d78f5495d53fc1f4f93e
2152 F20110216_AABTGI sachs_b_Page_020.txt
5b92b8333580b45de9c77b49addb05bb
88a5bc7bf6ea389479559ffafc6d32dc2e2175ab
208 F20110216_AABTFU sachs_b_Page_003.txt
66104542cad3cb07208c81d439c2a5aa
b49dd926c2ec98146a81928683e12e7d030ac29e
2203 F20110216_AABTGJ sachs_b_Page_021.txt
c3050574fbc2b55b145a0255fb65bc7f
c6916f2719de6c37b1600965be011549d8c4967d
369 F20110216_AABTFV sachs_b_Page_004.txt
9ca75255ffe31bcc7f3ade52a578f44d
637f4e19863d72210d0cd303ffc72d5796a3bbbd
2096 F20110216_AABTGK sachs_b_Page_022.txt
e35b3b6e5e494e16beeed0592bc5aeab
c230a0001367a159661a490f8f63a1a0312f49b0
3639 F20110216_AABTFW sachs_b_Page_005.txt
bff871e423370e8da39e02f31c3e867d
41c3f7ba815d64bb380356048d47488faa6b33b8
2242 F20110216_AABTGL sachs_b_Page_024.txt
33866a9eff6ec7da2c08e64221877652
c51de73b1a578117a06dc6c6ac6247bfa8a87c4d
2052 F20110216_AABTFX sachs_b_Page_006.txt
b4555300516b40255a33c4b0e4e6ce60
03843865255f91906c6887d48f19106076499e99
2169 F20110216_AABTHA sachs_b_Page_040.txt
0f2d6751d3330c8dac63d5c88c96bb10
c830d1d0b4c2e425ad9db7d958666cf2a58c4ebe
2150 F20110216_AABTGM sachs_b_Page_025.txt
d115f2e11417b912fc8cb2652d1f78a7
609f2ef723fd66f9d0f821d294d51374ac2a9c19
1630 F20110216_AABTFY sachs_b_Page_007.txt
5a6ce078632b2db395f942a2d887fe38
a5a8cda280a6efe05824c2a7cd495b5d419adec6
2138 F20110216_AABTGN sachs_b_Page_026.txt
792f1b933310799ab91110de1035eecb
ed87390b9de2c3f4e1fb0fd7417e4b8e942f9216
914 F20110216_AABTFZ sachs_b_Page_008.txt
5d7139f0760c30a111929cc73b30730e
72dca7591e191bf78606a2878e12406fbe11bc64
2090 F20110216_AABTHB sachs_b_Page_043.txt
dfaaab4ed00a84dc43ec550f5f77d4c9
3ef5590c49662efce99a9c8517343b3243a9447f
2216 F20110216_AABTGO sachs_b_Page_027.txt
97a1ab80f9127f3a299a6f5cab61ff60
8fe4a736e040d7fd35b5b172c84a3e9d7d86123f
2141 F20110216_AABTHC sachs_b_Page_044.txt
2138706b53d224b9b1891094b0720aa8
1f683e647eded36747944b2f2309daee6e7fd489
2213 F20110216_AABTGP sachs_b_Page_029.txt
0085a0b00dd601e5e6b5f444b69725cb
628645bbd5a43f1c4382ebe973bcf67b8a4bae02
2206 F20110216_AABTHD sachs_b_Page_045.txt
9067543ac9715c68f052f88f1d24fbb9
5a5e6b688214c9265ab6debdb08844c922ea958a
2179 F20110216_AABTGQ sachs_b_Page_030.txt
3450f3b7035f10da2f693ff13ae8aaad
777b8964e03e0b73410898316b9487b652eebb12
2103 F20110216_AABTHE sachs_b_Page_046.txt
ee7e1cdb403fe7d9ae00857585914a5b
d94fe1af0a74be9c3be2731f6ea91b5ef7c31f31
2155 F20110216_AABTGR sachs_b_Page_031.txt
1791566cb9052967a9fb339d91668b6d
795b2ca06886b926d19a4fd681eedfcde29e2e91
2091 F20110216_AABTHF sachs_b_Page_047.txt
e863b75083fe7a6c2c6cfed92227681f
01426cf8878e93d4eb7358fee2ef20a1031c21c0
2098 F20110216_AABTGS sachs_b_Page_032.txt
ea128b82106489ea809ad1a1d4d780a6
1b7d760ebe5dc5caa4113a7eb5cdff4e450c2387
F20110216_AABTHG sachs_b_Page_048.txt
80ed6e355dc688cdcc466daaac3412cd
5c26ac1e9ea66d818803d7c32a591c3d3f4d2dc9
2136 F20110216_AABTGT sachs_b_Page_033.txt
549240686d60a525f5d63b47f3c43901
a174d0de2d59aae76f6fff5965de57c8111ce13e
1659 F20110216_AABTHH sachs_b_Page_049.txt
05a80e9df6bccd3537a6dd74266ce25c
0f19891aef583f65aa782708909ab9d62b809ad2
335 F20110216_AABTGU sachs_b_Page_034.txt
ea8d06eb3e2a87d53941aa89625df73f
ba70412cb7ada18a2748575f8951306b5625019e
680 F20110216_AABTHI sachs_b_Page_050.txt
97eb8089921729e82764dff403e8d0f3
66f08ef3e38063b65d4a9c9db6d5c203d2d7579d
1911 F20110216_AABTGV sachs_b_Page_035.txt
9d95f277872ade6a1fcd7c1a6bfe1541
176952df62a901128e81b4f5600370c70e9c8876
1998 F20110216_AABTHJ sachs_b_Page_051.txt
3297fb8f2ed262b55a3594f6236abe95
50a3dee98e7ccc725c6cfdc2bfb5739780584183
1786 F20110216_AABTGW sachs_b_Page_036.txt
4e5578ae61bb5c8b44b7171195f57faa
68128099a0de69c5894b6160b851406820e204f0
F20110216_AABTHK sachs_b_Page_052.txt
801d5a993844afdbc9ce49cc81c08198
8b4b64d8c7d336d254d88fce5b4e24aba972af18
2183 F20110216_AABTGX sachs_b_Page_037.txt
151a9ce708a1af2d9bb7ec437cae458a
ea8740c07e14735beda5a7427cbe3fbd67ad22a9
2093 F20110216_AABTIA sachs_b_Page_070.txt
2c3290900f6fcc33dc740a647c935f7f
0437e6be9124c76ccf3c43a40a6394fe3907b49e
2156 F20110216_AABTHL sachs_b_Page_053.txt
519e3369f706b0e14e0bea562c11009a
99dea116827d713ed219de43e9a9e6e516fa13cc
2151 F20110216_AABTGY sachs_b_Page_038.txt
19b9ee008ed28b4e5839029c61e9caed
c8d50c3a642afbcde61592d711b044ea5b38dde8
2127 F20110216_AABTIB sachs_b_Page_071.txt
a6170854d08b93d0d18624c4ae912d39
5a38debd99c06cf2102b31569451132e8eaf04c3
2165 F20110216_AABTHM sachs_b_Page_054.txt
8d519036c8baf1f712d0751470fb5439
f52a7d7c53dba9e491319f62f16f4dbd37148cb7
2192 F20110216_AABTGZ sachs_b_Page_039.txt
9476c1e894c571deed1b736e288c73e1
bb9de5cfa4e32b78ff26c751d32c0d36406ef101
2109 F20110216_AABTHN sachs_b_Page_055.txt
9122b750fc29c2c707d2edb2e2414094
8efb7829cb664fb8f3166477c3412402558a9bdc
2145 F20110216_AABTIC sachs_b_Page_072.txt
c7bbf3c166bdc96d3cef00ca7c49464a
aeecb7e859ba967d58e36e22330e76a0a159c2ba
2167 F20110216_AABTHO sachs_b_Page_056.txt
c5b48624a98e5f705085f9d60aa195ba
5cd0f2531bf1667725c63776e2656e4c27b0e863
2117 F20110216_AABTID sachs_b_Page_073.txt
4935b8a44d65f92fa054ab25b9f99b21
98f56b02c7a423fa74bf0b9ebbd5f16babf8e5bf
2230 F20110216_AABTHP sachs_b_Page_057.txt
0d0a764dd09233db1d27c1a3cfe0503f
b2fc9371f019d85432a0b72b256c4e08d22df628
2153 F20110216_AABTIE sachs_b_Page_074.txt
451166060e6d722dbda52f87e329e4a3
017df883f930a19d97359ea487d3cc99555a5444
1605 F20110216_AABTHQ sachs_b_Page_059.txt
5341dd222c13b516e3546b198c0109f8
08fe6dc4f4e4b435c06b4d2b8dee5f20019e90f6
2226 F20110216_AABTIF sachs_b_Page_075.txt
3f7762c0139d9541ec7b18b7be277467
a6b39c978026c551cdfffda84d61b9ca9080c8ae
F20110216_AABTHR sachs_b_Page_060.txt
3030cf7069f1170d9a1981e4f1fb0d2e
4dcd4b8b1b9f40c498819989566164fd90017a0a
2163 F20110216_AABTIG sachs_b_Page_076.txt
b0bba055d41d3b8e05cb7950bb76b28c
319decaa3c8c8acbfca541bfa7b7d37bc0d55e92
1675 F20110216_AABTHS sachs_b_Page_061.txt
94771f338129f690cfbd638e3da128db
5d3669047bae53d7741f798ea502c1e3209a3b16
F20110216_AABTIH sachs_b_Page_077.txt
a76e5946e2fe011fb6d325f5f629a4cd
4aeece13d3e2cc27ebfc6362a7a948a77c8fc365
2361 F20110216_AABTHT sachs_b_Page_062.txt
e4e7292745869e4828318f2afd2edffb
b7f6733a0da4ced6750f469a771d482870cbcd04
2158 F20110216_AABTII sachs_b_Page_078.txt
e1d2adeb341c6510cf6bd34f6c12cc88
004808e98e75e09706994471d5d35403fb0279bb
1154 F20110216_AABTHU sachs_b_Page_063.txt
2c61c0d1963464b790246c7aa2e0033c
2d05563e16baa9041d8c796a6216730309d426e3
2214 F20110216_AABTIJ sachs_b_Page_079.txt
95418bba172990d806aafa7933af6020
8f4a9106f24719f1dd12a85c6c04b75cfb873577
588 F20110216_AABTHV sachs_b_Page_065.txt
00aae6a6e0e3f2b84f884cd9a20a79c9
1bcfa7b815896a7e5cadee220c41b2bd5cfe9660
2234 F20110216_AABTIK sachs_b_Page_080.txt
dac21ba0a063e493e19df58bbf0d7dfe
08fe68e75b373c41f8ea0f1fd2fed201d3cb351c
352 F20110216_AABTHW sachs_b_Page_066.txt
e0496b3d4a0215504b506e1d1e5f714a
b44275c891733416c330cb84516390cf52789aab
2773 F20110216_AABTJA sachs_b_Page_099.txt
c5786392f6f38268e922028f8131178d
f45f78bec1b8fa90006de239d998c25284ca941b
2236 F20110216_AABTIL sachs_b_Page_081.txt
2e4473daa9f8bcf0c6439bcf4081128c
339a260792d1076259c3d49e54553899c62e0567
2130 F20110216_AABTHX sachs_b_Page_067.txt
e06ae1639e6249c2eaf1ecc4ef022bf3
ee1258367621b59258fdc87d1040c76c62be4230
2401 F20110216_AABTJB sachs_b_Page_100.txt
e8da27862af6de41442a456f7e00e440
e68012be332cd73acb134d6518beea1b1b4b1533
2060 F20110216_AABTIM sachs_b_Page_083.txt
931ffd3954a5033a57074930cba46cca
ee2c9994b43170edf731538a86abf360fc9c7b2a
2229 F20110216_AABTHY sachs_b_Page_068.txt
fc5ee2b1fc7e90c15f88f8e5065dcf51
e6b2ac9231951e32f88be84a87e24843c5d52ab8
2623 F20110216_AABTJC sachs_b_Page_101.txt
77bc900e202e19f98bd1dd8993801517
2cd42da3d0a2255af8ce37b38a6119f2d20983fb
2089 F20110216_AABTIN sachs_b_Page_084.txt
a07d706607d03aa1b880a2166ca963ed
77817fd6f128c88b8fbe061a5796ac7550c65c5c
2055 F20110216_AABTHZ sachs_b_Page_069.txt
a9cddfbde14becebecce4cf5b503fdbb
87c3dd37f9d021001e693476f6d36040599d45ab
F20110216_AABTIO sachs_b_Page_085.txt
0f3a637d76d1302d5d932f240feb7287
27db725a2131e17df04002ab71bb7d81432be2bc
2684 F20110216_AABTJD sachs_b_Page_102.txt
f009b0dd437eb1eb181e0bfa8921d8c0
7c4213bdba8354096e8e2496b00f03f5160e1cfc
2287 F20110216_AABTIP sachs_b_Page_086.txt
ad6338b65632c7abb434675f05a6ef42
2eeebde1be4ca0dd11ef2f412dd920c78ce49ce3
2559 F20110216_AABTJE sachs_b_Page_104.txt
5befe27b6b8fdfe4f63429c9270e8510
b8c0fa8dee404fa188d91100b2efd8afef25641f
2228 F20110216_AABTIQ sachs_b_Page_087.txt
1bcbe5c8341bb1e52681d89a35d43772
06385f76f5ddb11990028a53cc3894e4a46bdc83
2661 F20110216_AABTJF sachs_b_Page_105.txt
50e3810f545d60f2565f8b4e781a7273
e6037ebc9735035de6ed7acbe9b0c9324d9f549c
2104 F20110216_AABTIR sachs_b_Page_089.txt
f5bdf29472fb49c60c541c2235779c8c
080e0ca2f4394900ff2e9b61d237af69fbe604d0
2460 F20110216_AABTJG sachs_b_Page_106.txt
f8f65a9b9835601d16f35a6b4208b9ba
192c05bfad37aae24a3140e54f4b23bd59c9cba9
2128 F20110216_AABTIS sachs_b_Page_091.txt
d7f23a1dc9ea9e24cf85d0da468b529c
739b61fd73c75a3c5d8f1d0fc46bbadb8d371233
1418 F20110216_AABTJH sachs_b_Page_108.txt
cf33ea776d12318485fe5d3faa2a37d7
ed83f8421cc1188cd8f6da5bb960329194b07a8e
2245 F20110216_AABTIT sachs_b_Page_092.txt
05c619133fcfca8f8bb7cc768d6d1dc6
621070b68b503b9ef753896eadd957670b8c1224
1062 F20110216_AABTJI sachs_b_Page_109.txt
f3d5a64e0bb8e0eacdb2bda6d187cbd7
15a435147db7b971288c2bd996aa9a88ea7e2064
848 F20110216_AABTIU sachs_b_Page_093.txt
81b58fce1eb49effc37b182f2e56ce4a
206e8ae16ccb6078fca8591835f173e3f0c547cc
4087 F20110216_AABTIV sachs_b_Page_094.txt
e99f93bacacf9611656442d94241074f
96d17622d86dc0ac2cdd86a69aa2b23e2758e024
10414 F20110216_AABTJJ sachs_b_Page_001.pro
fb0dde6eaac8a006f64a5f0b8c787bd5
40c9aafcdc6bd8922c4548a55f319e762d4af759
567 F20110216_AABTIW sachs_b_Page_095.txt
274bd0e2e320c88c3e063a1ced6a230f
2d96b5645005d4bc58fc734d597758db8c20c2ee
989 F20110216_AABTJK sachs_b_Page_002.pro
132859b9fb553009a9aeae32b8a1a65a
d8db24679a4401016ed7376d5ae9fae8a8b867e0
F20110216_AABTIX sachs_b_Page_096.txt
dd7268dc4b48d6278c3bc88c8c3ba372
a6f6157bbb69a3e1b0eeb53c9d40528c2f91c43e
54289 F20110216_AABTKA sachs_b_Page_026.pro
7db2ee72723ee4caaf9976cba2e37395
235519db6328797ad90e016df3463ae24185eefa
3394 F20110216_AABTJL sachs_b_Page_003.pro
dc125921fa177dbaa6e8e78bb2e71e4b
abf37be704d76e155f164be080f9fc9d833e49d2
2376 F20110216_AABTIY sachs_b_Page_097.txt
02685475a7bbb0c2af7bcbec74f31be3
bb3674f034fcedb2e4f318e3e07ca7af15cb81fb
56435 F20110216_AABTKB sachs_b_Page_027.pro
b86eb7e3da270ead22e02ce11a1e5a3a
3bde13517137b0dea5867a7affd14cb664bbead4
40469 F20110216_AABTJM sachs_b_Page_007.pro
14ee4358adae7b2a0b0d90dc56dcd4c8
ae62711d08dd0fc287bc1018ef16b4f7ef670e1a
2347 F20110216_AABTIZ sachs_b_Page_098.txt
2d7af4b610b1a1f7b60002d793adf21a
633b1b7408be396a5c134e2d74958926df8176cc
52939 F20110216_AABTKC sachs_b_Page_028.pro
3db97f9f1b70438f24b713e1c174f3a4
ed307052b40a7179febf1720dd6b41e213b955dc
21412 F20110216_AABTJN sachs_b_Page_008.pro
9e5268994744d6814094d17aeb13325d
4704aee5745448b93992dbba1007548f22993f1e
56308 F20110216_AABTKD sachs_b_Page_029.pro
05c49534fe4591be6570cd6992d556ef
c0f15bc1840f698a21d31bcb17f16b3a5e8f0889
42737 F20110216_AABTJO sachs_b_Page_009.pro
9cf879e34ad7c792dfc883c1bdc7d84f
20c855757e92a85d0e564faa937e20074db2a964
31343 F20110216_AABTJP sachs_b_Page_010.pro
efab57c6b65f0f76714e74d34ff0b8d1
94b5cb9923070edf85d09a6f9ff45f6f519b5bc7
55546 F20110216_AABTKE sachs_b_Page_030.pro
f0f56a138346e11f7b1f3dac5fa3c012
51851073b26ea046e9b100647111e6789d4d0fc0
51946 F20110216_AABTJQ sachs_b_Page_011.pro
06e0c25071b88491f197e9ea121898eb
72317435b3497fd8fdf6de7d73268bc89fe5966d
54839 F20110216_AABTKF sachs_b_Page_031.pro
7ce64aea007c628903bb31f7c90159c9
5d0a9eefc8ad4a6b0c6087b1a83814036433ce1e
54004 F20110216_AABTJR sachs_b_Page_012.pro
898c0fd1c00f9f9285920b757928fc6d
0600d185e7cf29d804932f85003dd97185f8c58e
53325 F20110216_AABTKG sachs_b_Page_032.pro
f63973b0185891287a59d1a6e9479f8c
97ab53a33ee469ec21949ada6d28629e6905ec1d
58929 F20110216_AABTJS sachs_b_Page_014.pro
e7446857b34c27d65b876b9f00289da1
beeccd09fe0e71bb30d311951104cf95666480f4
52954 F20110216_AABTKH sachs_b_Page_033.pro
618ceaf75753bf570ecedc995b06859a
ba839b0af660d05f460e8e4a10f18f36e3a6005e
56800 F20110216_AABTJT sachs_b_Page_015.pro
42f86957735747688a16196045f43b84
17ebbad400eccd1b88be164f739ba9383f42dab2
8364 F20110216_AABTKI sachs_b_Page_034.pro
492c77d3b4d23307003ef583b653ba6e
9dd7a0b7461a93162923983ef0f0d955154de34f
55879 F20110216_AABTJU sachs_b_Page_016.pro
825d3079891b1141454d121ddd3fec40
7cee2317720cda19fed5c70de41cf682528f5326
42838 F20110216_AABTKJ sachs_b_Page_036.pro
d101c9a12718fafa49ab9bd816107e8d
9f63c67fb4ff6874c05956c90c4113cf0e0164c8
54239 F20110216_AABTJV sachs_b_Page_019.pro
b4122f64dfc4b13a97c5c85098cfafb5
1965ebb3ba03557715c63fa4bed994c9f61f78b5
52516 F20110216_AABTKK sachs_b_Page_037.pro
e54fb51220b4ba27dc35bb96d8723b2e
b1ef654d63d77749d9b1c861065deb93098d358c
56304 F20110216_AABTJW sachs_b_Page_021.pro
abc4ff9e09e0538ec3ceff64fb02f8ce
7451697565c3229053841eafae7f7f4ce6290b91
53990 F20110216_AABTLA sachs_b_Page_054.pro
fb88a383889fe1c535136e06ee3f5cd0
71c0079df2ee15cb72d38e047cc7f884d186de09
54790 F20110216_AABTKL sachs_b_Page_038.pro
5655857abc29556e97db37c2688e1e61
c73b3c0db048416844c23b235ff8c306cc0b0723
53161 F20110216_AABTJX sachs_b_Page_022.pro
348b0008119cb89cd420c9ee61491535
445136c69b4eccdc8183fe9198091a48b7992da9
53793 F20110216_AABTLB sachs_b_Page_056.pro
b70e9b8f40f2986deeaf9e313fc232f5
51c36c197f342eed72564b4e7eb7d3ca498c4b7a
55894 F20110216_AABTKM sachs_b_Page_039.pro
dacc998acac1d6a0f7e922b6d97160ab
efbd600aa62fce49470ad0702e9c9972c35540c1
57144 F20110216_AABTJY sachs_b_Page_024.pro
4ebb3735f791a2e8acba7965877860fb
6c56fb0539aefd09d8ec1e7faee148d7c1e340ef
56992 F20110216_AABTLC sachs_b_Page_057.pro
0bbd5660dc4bbf77c366b78eef3afde6
371fd4c9b7e5cc0552ff80b5c7b631fed1be6f2c
55257 F20110216_AABTKN sachs_b_Page_040.pro
f344eb15af81b825276487ce0ac86629
c2c6d7ce234c5311e5c425f9d1f06aa266674f03
54956 F20110216_AABTJZ sachs_b_Page_025.pro
f42daa1214ff70038aaa862966f0e828
117fbf2399294f3ae4cfc154a996bee65e7f9945
49411 F20110216_AABTLD sachs_b_Page_058.pro
b1cb49c55547f090dfc11f980006199c
c1f0e6d00f45d340694adbc09dc832d66d56b6c7
51673 F20110216_AABTKO sachs_b_Page_041.pro
9d2d11370ef79015d1af98c85d4fc389
c13580120b5890784fe712fcc5a9917637e4dd7a
40063 F20110216_AABTLE sachs_b_Page_060.pro
18635842f090c81a9aea14b2f8f24d3d
b36cf72972405a7b44948d0acb225c4dac52dd59
51132 F20110216_AABTKP sachs_b_Page_042.pro
23130bc84ad51723c6c18683cb9d93a5
52fa2e4a80bcd94982ae205cdde7ea221adce5a4
51941 F20110216_AABTKQ sachs_b_Page_043.pro
311f8142000e7c5cce6c5e5a5e2c78e1
3fecbd51c12e9fde34eaaf3b09accfe46922190e
33087 F20110216_AABTLF sachs_b_Page_061.pro
4d604bb76dafa00ddeb2b96de125300b
cca4340a9fcbca82523ff9850d1cf83e4566b562
54536 F20110216_AABTKR sachs_b_Page_044.pro
a30c3891ecafaedd7d0b60ca10740c2f
57d444a0a0714a2ce71f130dcdda769a87ee853a
47610 F20110216_AABTLG sachs_b_Page_062.pro
4a6335c55fe675cfaa02a464c68a42cf
7668da47d05a190b5812fc7416aff60a45c1dbee
56226 F20110216_AABTKS sachs_b_Page_045.pro
a727cd4db67bdf5a15ebec62bf5de166
754645b61d6770509fa10149c35e3943b6daeec1








THE DIAGNOSTIC UTILITY OF A MULTI-TASK VERBAL FLUENCY PARADIGM IN
FRONTAL AND TEMPORAL LOBE EPILEPSY: AN ANALYSIS OF FLUENCY TYPE AND
QUALITATIVE PERFORMANCE




















By

BONNIE COLLEEN SACHS


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

2009



































2009 Bonnie Colleen Sachs































To my chair, Dr. Russell M. Bauer, and all of my other mentors who have helped me navigate
the world of neuropsychology









ACKNOWLEDGMENTS

I thank my family for their support and encouragement during graduate school and

internship. I also thank the members of my dissertation committee: Russell Bauer, Ph.D.; David

Janicke, Ph.D.; David Loring, Ph.D.; and Steven Roper, M.D.; for their thoughtful contributions

to this project.










TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ............................................................................................... ............... 4

L IST O F T A B L E S ... .. .................. ..... ..................................................................... .............. ....... 7

L IST O F F IG U R E S ................................................................. 8

A B S T R A C T ....................................................................................... ......... ...... 9

CHAPTER

1 BACKGROUND AND SIGNIFICANCE ........................................................................... 11

Frontal and Temporal Lobe Epilepsy................................................ ........ ... ........... 11
Measurement of Neuropsychological Functioning In FLE and TLE ................................... 14
V erbal F lu ency ...................................................... 17
Semantic Fluency............................................................. 18
Phonem ic Fluency ............................................ ................... .................. ...... 20
Novel Techniques for Dissociating Frontal and Temporal Lobe Impairment ........................23
A ction/V erb R etrieval .................................................... ...................... ....................... 23
Proper N am e R etrieval ........................................................................ 28
Q u alitativ e A n aly sis ............... ............................................................... .............. ....... 3 0
Su m m ary ..................................................................... ........... . 3 3

2 SPECIFIC AIMS & HYPOTHESES OF CURRENT STUDY ............................................35

Aim 1.....................3 5
A im 2 ................ .......................... ................. .......... 3 5
Aim ........................................ 35
A i m 3 ............................................................................................................................................ 3 6

Aim 4................. ...................... 36

3 SU B JE C T S & M E TH O D S ............................................................................. 37

Study Participants....................... ................ ....................... 37
E p ile p sy P a tien ts ............................................................................................................ 3 9
P re -su rg ical p atien ts ............................................................................................... 3 9
P ost-surgical patients .................................................... 4 1
H healthy controls ..................................................... 42
Measures ........................................... ............... 43
Pre-surgical Patients ........................................ 43
A additional M easures..... ..................................... ........................................ ............. .. 46
Post-surgical Patients and Healthy Controls........................... ....................................48






5









4 R E S U L T S ............ .............. ...................................................................................................... 5 1

A im 1 ............................................................................................................................................ 5 1
Aim 2...................................................................... 53
A im 3 ........................................ ..... ............... 54
A im 4 ........................................................................... ...... .... 5 6
A additional Study A im s .................................................................. 58

5 D IS C U S S IO N ...................................................................................................... ..................... 6 7

Summary of Findings .............. ..... ......... .... ...............67
Interpretation of Findings ..................................................... 72
Semantic and Phonemic Fluency .............. ......... .............72
A action Fluency ......... ............ .........................................76
N am e F lu en cy .............................................................. 80
Qualitative Analysis of Fluency Performance .................................. 83
Limitations of the Present Study ......................................... .................. ......... 86
Directions for Future Research and Clinical Use ..... ................ ............... 89

APPENDIX

A STANDARD NEUROPSYCHOLOGICAL TEST BATTERY (SNB) ............... ...............94

B SAMPLE RESPONSES FROM FLUENCY DATA.............................................. 95

R E F E R E N C E S ................................................................................................................................... 9 7

BIOGRAPHICAL SKETCH ......................................... 109









LIST OF TABLES


Table page

3-1 P patient seizure characteristics................................................................................. ..... 49

3-2 D em graphic and clinical characteristics ........................................ ......................... .. 50

4-1 Perform ance on fluency m measures ................................................. ........................... 60

4-2 Multivariate analysis of fluency performance in patients.................................................. 60

4-3 Four fluencies predicting patient group membership ................................... ........ ... 60

4-4 Semantic and name fluencies predicting patient group membership..............................60

4-5 Correlations coefficients for fluency measures (patients only) ...................................60

4-6 Correlations coefficients for fluency measures (patients and controls)............................61

4-7 Correlations amongst neuropsychological measures (patients only)..............................61

4-8 Correlations amongst neuropsychological measures (patients & controls)........................62

4-9 Perform ance on nam ing m measures ................................................. ........................... 62

4-10 Correlations between naming and fluency measures (patients and controls)..................... 62

4-11 Correlations between naming and fluency measures (patients only)..............................63









LIST OF FIGURES

Figure page

4-1 Overall fluency performance across groups.......................................................... 63

4-2 Receiver operating characteristic (ROC) curve for semantic and name fluencies
prediction g p patient group p ............... ............................................ .......................................... 6 4

4-3 M ean num ber of clusters by group.................................................... ............................ 65

4-4 M ean num ber of sw itches by group............................................. .................................. 65

4-5 M ean clu ster size by group ........................................................................... ........ .......... 66









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


THE DIAGNOSTIC UTILITY OF A MULTI-TASK VERBAL FLUENCY PARADIGM IN
FRONTAL AND TEMPORAL LOBE EPILEPSY: AN ANALYSIS OF FLUENCY TYPE AND
QUALITATIVE PERFORMANCE


By

Bonnie Colleen Sachs

August 2009

Chair: Name: Russell M. Bauer
Major: Psychology

Epilepsy is a chronic, disabling condition affecting roughly 50 out of every 100,000

Americans. The most common site for seizure onset is in either the frontal (FLE) or temporal

lobes (TLE) of the brain. Identification of the onset of seizures is important, and partly

determined by cognitive test scores. Elucidating a pattern of neuropsychological test

performance in FLE and TLE is complicated, but allows for more accurate identification of

seizure onset, which is essential in treatment planning. Presently, these cognitive patterns are

poorly defined and overlapping, partly due to the lack of specificity of our cognitive tests. The

purpose of this study was to evaluate the utility of multiple measures of verbal fluency in the

differential diagnosis of intractable FL and TL epilepsy.

Patients in the study included pre- and post-surgical refractory epilepsy patients with either

temporal (N=14) or frontal lobe epilepsy (N=7) and healthy age and education matched controls

(N=20). Patients completed a battery of neuropsychological tests thought to be sensitive to FL

and TL functioning, including standard semantic and phonemic fluencies and novel action and

name fluencies.









We found significant differences between patients and controls for action and name

fluency. Patient groups were not statistically different on fluency measures, though effect sizes

indicated FL patients outperformed TL patients on name fluency. A qualitative analysis of

fluency (clusters and switches) only differed for patient groups on name fluency as well. Only

name and semantic fluency were adequate predictors of patient group membership. We found

support for the notion that all fluency measures were related to overall intellectual ability and

verbal/semantic factors. However, measures of semantic and name fluency were more related to

semantic abilities and phonemic and action fluency were also related to measures of executive

functioning.

Results of the study indicate that in a mixed pre- and post-surgical epilepsy population,

phonemic, action, and semantic fluency were not specific to frontal and temporal lobe

functioning. Further, qualitative assessments of fluency did not offer significant information

about seizure foci. Name fluency differentiated well between patient groups and appears to be a

novel measure sensitive to the integrity of the left temporal lobe.









CHAPTER 1
BACKGROUND AND SIGNIFICANCE

The purpose of this study was to evaluate the utility of multiple measures of verbal

fluency in the differential diagnosis of intractable epilepsy of temporal versus frontal lobe origin.

The following sections in the present chapter provide background on the diseases of frontal and

temporal lobe epilepsy, and the types of cognitive and neuropsychological deficits that are most

common in these populations. The standard neuropsychological assessments used to detect these

impairments are also reviewed, including the limitations of these measures in patients with focal

epilepsy. Subsequently, the literature positing material-specific impairments in the production of

"action words" and "proper names" in patients with frontal and temporal lobe disease is

reviewed in order to substantiate the incorporation of these features into traditional

neuropsychological assessment. Finally, specific aims and hypotheses for this study are

presented.

Frontal and Temporal Lobe Epilepsy

Epilepsy is a chronic, often disabling disorder affecting between 30 and 60 persons per

100,000 in the United States (Hauser, Annegers, Rocca, 1996). In other countries, prevalence

ranges between 30 and 100 cases per 100,000 individuals (Forsgren, Beghi, Oun, & Sillanpaa,

2005). While epilepsy accounts for only 5-10% of all disabilities in the United States, the

consequences of the disorder can be quite severe, including impairments in quality of life,

inability to drive and associated loss of independence, less frequent social interaction and lower

marriage rates (Sperling, 2004). Further, uncontrolled seizures can lead to neuronal death and

physiological dysfunction. While many individuals with epilepsy are diagnosed at a young age,

epilepsy can begin at any point in the lifespan (Paradowski & Zagradjek, 2005; Hauser,

Annegers, Rocca, 1996). Often, the etiology of the disorder is unknown. In some cases, however,









its development has been linked to neoplasm, head injury, family history of seizures, or severe

illnesses such as meningitis or encephalitis (Chin, Neville, & Scott, 2005; Wakamoto, et al.,

2004; Annegers, Hauser, Coan, et al., 1998; Davies, Hermann, Dohan & Wyler, 1996).

In recent years, there have been great advances in treatment options for individuals

suffering from this disorder. Currently, there are dozens of pharmacological agents that can be

used to prevent seizures, or to help lessen the frequency or intensity of the epileptic events (see

Perucca, 2005 for a review). While drug therapies are often effective in managing epilepsy,

many cases remain refractory to pharmacologic interventions. Individuals with medication

refractory epilepsy, many of whom have complex-partial (or focal) epilepsy, often suffer from

years of debilitating and dangerous seizures.

The most common types of focal epilepsies arise from either the temporal lobe or frontal

lobe (Engel, 1996). Despite the relative number of patients with frontal lobe epilepsy (FLE),

temporal lobe epilepsy (TLE) patients may become candidates for seizure surgery more

frequently due to the difficulty in diagnosing and localizing frontal lobe epilepsy. This challenge

is accounted for by several different reasons: FLE may be associated with diverse seizure

semiologies and EEG recordings in FLE often show widespread epileptic activity, and

neuropsychological profiles of patients with suspected FLE are often not distinct from patients

with other epilepsies, including the most common variety, TLE (Exner et al., 2002; Hermann,

Wyler, & Richey, 1988; Helmstaedter, 1996).

Although no signature pattern of impairment exists and much controversy exists around

the "executive functions" of the frontal lobes, behavioral and cognitive characteristics of frontal

lobe epilepsy, and frontal lobe damage in general, have been described in the literature. While

executive functions are complex and multi-faceted, most of the literature concerning executive









dysfunction does implicate frontal lobe damage as etiologic. The heterogeneous term, "executive

function" refers to both the notion of cognitive control over mental abilities, and the ability to be

adaptive and flexible in novel or unpredictable situations, which is clearly important for

everyday functioning. Generally, executive functions can be thought of as behaviors that include

problem solving, planning, abstraction, response inhibition, self-awareness, cognitive flexibility,

cognitive control and hypothesis generation (Risberg & Grafman, 2006; Lezak et al., 2004).

Patients with frontal damage can exhibit various types of executive dysfunction on

neuropsychological tests, depending on which sector of the frontal lobe is involved. Deficits

include impairments in planning, initiative, inhibition, behavioral control, emotional regulation

and working memory (Helmstaedter, 2001).

In addition to cognitive deficits, patients with lesions to the frontal lobe can exhibit a

variety of characteristic personality, behavioral, and emotional changes. Damage to the

orbitofrontal region, for instance, can cause behavioral disinhibition, emotional liability,

impulsivity, altered social conduct (so-called acquiredd sociopathy"), and changes in personality,

whereas damage to the lateral prefrontal cortex can cause a decline in working memory abilities,

impairments in abstract reasoning, mental inflexibility, and difficulties with decision making

(Bechara, Damasio, & Damasio, 2000; Tranel, 1992). Many patients with damage to the frontal

lobe display unique characteristics on neuropsychological and behavioral tests including

perseverations (i.e. the inability to stop a behavior/response), intrusions, motor impersistence

(i.e. an inability to sustain a motor gesture or action over a period of time), and poor self-

regulation (Alvarez & Emory, 2006; Helmstaedter, 2001). In addition, depending on the site of

the frontal lobe lesion, patients can also display difficulties with expressive speech, motor

weakness or incoordination, apathy, or problems with abstract thinking.









While many of these characteristics are unique to persons with damage to the frontal

lobes, others are found commonly in other patient groups. This is the certainly the case when

studying patients with frontal and temporal lobe epilepsy. Typically, patients with TLE localized

to the left hemisphere (LTLE) display robust reliable neuropsychological impairments that are

generally replicated across studies. These include deficits on tests on object and person naming,

verbal comprehension, and learning and memory for both simple and complex forms of verbal

material (Hermann, Seidenberg, Schoenfeld, & Davies, 1997; Hermann et al., 1999). However,

LTLE patients commonly exhibit difficulty not only with verbal learning and memory, but also

with response inhibition, impulsivity, set loss, and difficulties with mental flexibility and abstract

thinking. These overlapping patterns of impairments on neuropsychological tests are likely due

to a variety of factors, including the fact that patients with TLE often have propagation of

abnormal electrical activity from the temporal to frontal regions, thereby causing potential

impairment (Hermann & Seidenberg, 1995). Further, patients with TLE can exhibit reductions

in white matter volume in frontal cortex, in addition to a reduction in overall cerebral volume

and gray matter changes (Hermann et al., 2003; McMillan et al., 2004; Oyegbile et al., 2006). In

addition, patients with longstanding seizure disorders may exhibit depressed cognitive profiles

on multiple cognitive domains due to the cumulative effect of uncontrolled seizures (Jokeit &

Ebner, 2002). Further, this lack of differentiation between frontal and temporal lobe epilepsy

patients may also reflect the relatively poor specificity of our assessments for localized brain

dysfunction. Given the relative overlap of these cognitive profiles, identification of seizure

localization is at best a complicated and ambiguous task for neuropsychologists.

Measurement of Neuropsychological Functioning In FLE and TLE

As previously mentioned, patients with frontal and temporal lobe epilepsy may present

with overlapping neuropsychological profiles. This may result, in part, from overlapping neural









pathology, and in part because of relatively poor ability of clinical tests to discriminate frontal

from temporal lobe epilepsy. A pre-surgical neuropsychological test battery for epilepsy patients

includes tests of overall cognitive functioning, expressive and receptive language, verbal and

nonverbal memory, processing speed, visuoconstructional ability, attention, "executive

function", and mood. Tests routinely used to assess for "executive" or frontal lobe dysfunction

usually include measures such as the Wisconsin Card Sorting Test (WCST), Stroop Color-Word

Test (Stroop), Trail Making Test (TMT ), and measures of verbal fluency, including phonemic

and semantic fluency tests (Lezak et al., 2004). Additional measures such as Luria's Motor

Sequencing Tests (Luria, 1966) may also be administered in some clinics (Stuss & Levine,

2002). Combined results on these measures is then used to determine whether or not a patient

displays significant executive dysfunction and, in combination with other test results, whether or

not the patient displays a localizing pattern of deficits. The difficulty in clinical decision-making

lies in the fact that these tests are sensitive to the presence of brain disease or damage, but may

not be sufficiently specific to damage in a particular region of the brain.

A body of evidence supports the notion that impairment on measures of "executive

function" are not specific to damage of the frontal lobe, and do not discriminate frontal patients

from those with temporal or other types of damage. In epilepsy populations, Hermann, Wyler,

and Richey (1988) initially documented significant errors in planning and problem-solving on

the Wisconsin Card Sorting Test (WCST) in a small group of left and right TLE patients

compared to generalized seizure patients and controls. The finding of poor performance on the

WCST in TLE, as measured largely by the number of perseverative errors, has been confirmed

by others (Martin et al., 2000; Trennery & Jack, 1994; Corcoran & Upton, 1993). A recent meta-

analytic review (Emory & Alvarez, 2006) found that out of twenty-five lesion studies, twelve









found that adults with frontal lobe lesions performed more poorly on the WCST than healthy

controls, and 10/16 studies suggested that frontal lobe lesioned persons performed worse than

those with extrafrontal lesions. However, they also reported that two studies found no differences

between frontal lobe patients when compared to normative data, and nine studies found no

significant differences between groups with frontal cortical lesions and those with lesions

elsewhere in the brain (i.e. basal ganglia, diffuse lesions).

Additional studies using the Stroop paradigm have found that non-frontal epilepsy

patients perform poorly on this measure. One recent study by McDonald and colleagues

(McDonald et al., 2005) found that patients with left lateralized TLE were significantly impaired

on measures of switching and inhibition on the Color-Word Interference Test (McDonald et al.,

2005). Despite the frequency of use by neuropsychologists, few lesion studies have employed the

Stroop paradigm to examine its specificity to frontal lobe lesions. Of those that have, only two

studies found that persons with lesions to the frontal lobes performed worse than controls (Stuss

et al., 2001; Vendrell et al., 1995), and two studies reported those with frontal lesions perform

worse than nonfrontal controls on the interference trial (Perret, 1974; Stuss et al., 2001). On the

other hand, studies have found the opposite pattern; frontal lobe and temporal lobe lesioned

patients performed equally poorly on this measure (Blenner, 1993).

Demakis (2004) conducted a meta-analysis of studies which have employed the TMT

and Category Test to determine the relative utility of these instruments in detecting frontal lobe

damage or dysfunction. Based on the studies included, 321 participants were included in the

overall meta-analysis, although sample size varied slightly by test examined. Surprisingly, the

results of the study indicated that frontal patients performed significantly worse than non-frontal

patients on Trails A (thought to assess mainly psychomotor speed), but did not perform worse on









the Category Test or Trails B. Exner and colleagues (Exner et al., 2002) found that patients with

frontal and temporal lobe epilepsy had inferior Trails A and B performance compared to

controls, but did not significantly differ from each other on these measures.

In sum, these results and others indicate that tests commonly thought to assess the

integrity of the frontal lobe are in many cases, sensitive to frontal dysfunction. However, many

studies have failed to find such effects, and other studies have found that, while some of the tests

may be sensitive to frontal lobe damage, they are nonspecifically impaired in damage to the

temporal lobe and elsewhere in the brain. These results raise questions about the utility of many

so-called "frontal-executive" tests, and at best, indicate that such tests may be sufficiently

sensitive but insufficiently specific. These findings suggest the need to further develop

neuropsychological methods with greater specificity for frontal lobe disturbance in clinical

populations.

Verbal Fluency

Verbal fluency measures are amongst the most common measures administered in

traditional neuropsychological assessment (Stuss & Levine, 2002), as more than 50% of

neuropsychologists report using these measures in standard clinical practice (Butler et al., 1991).

Fluency measures are quickly administered, easily scored, and readily available. Further,

adequate norms exist for these measures making their use in clinical practice even more

prevalent. These measures require "time-restricted generation of multiple response alternatives

under constrained search conditions and involves associate exploration and retrieval of words"

(Henry & Crawford, 2004). Although a variety of fluency measures exist (i.e. written fluency,

figural fluency) the most common varieties of fluency tests are oral in nature, and assess word

generation to either a phonemic or semantic cue. Phonemic fluency requires generation of words









that begin with a particular letter (i.e. F, A, S, or C, F, L), whereas semantic fluency assesses

word production to a given category (i.e. animals, items in a supermarket, fruits and vegetables).

Semantic Fluency

Presently, there is disagreement in the literature as to the extent to which semantic

fluency is sensitive to the integrity of the frontal lobes. Some evidence suggests that phonemic

and semantic fluency may impose differing demands on frontal-executive processes; searching

for semantic items within a larger superordinate category places demands on well-established

search mechanisms that are congruent with organizational structures in our environment (e.g.

generating items that can all be found in a supermarket, as opposed to generating items by letter,

which contain no inherent semantic relationships). Perret (1974) has argued that because the

search criteria for semantic fluency are consistent with the natural organization of the human

lexicon, the demands of this task rely less on the executive processes, and more on the integrity

and organization of semantic memory stores. Others argue, however, that patients with executive

dysfunction are unable to perform effective and strategic searches through memory, irrespective

of whether the search is semantically or phonemically driven (Baldo et al., 2006; Baldo &

Shimamura, 1998; Troyer et al., 1998), and as such, would be equally impaired on semantic and

phonemic fluency tests.

Empirical data provides some clarity to the theoretical debate, although the body of

literature is not entirely consistent. For instance, Drane et al. (2006) found that patients with

frontal lobe epilepsy were more impaired than a group with temporal lobe epilepsy on measures

of semantic fluency. A study comparing patients with focal anterior and posterior lesions found

that both types of lesions produced impairments on semantic, or category, fluency (Stuss et al.,

1998). Additional studies employing a variety of populations have found similar impairments in

semantic fluency in frontal-lobe patients (Baldo & Shimamura, 1998; Costello & Warrington,









1989; Owen et al., 1990; Randolph et al., 1993). Conversely, studies report spared semantic

fluency in patients with frontal lesions (Corcoran & Upton, 1993; Joanet & Goulet, 1986; Jurado

et al., 2000, Vilkki & Holst, 1994). In addition, secondary interference tasks though to disrupt

frontal lobe functioning have not been successful in disrupting category fluency performance

(Mack, 1994; Martin et al., 1994).

Functional imaging studies have attempted to delineate brain regions associated with

successful category fluency performance. Mummery (Mummery et al., 1996) reported significant

left temporal lobe activation in the inferior and anteromedial regions while patients performed

category fluency tasks, but did not find significant frontal lobe activation for this task. Using

voxel-based lesion mapping, Baldo and colleagues (2006) found that category fluency

performance was associated with lesions in the left temporal lobe, post-central gyrus, parietal

cortex, and putamen. When examining areas specific to category fluency, the most important

regions of interest were in the temporal (Brodmann's Areas (BA) 22, 37, 38, 41, and 43) and

parietal cortices (BA 7, 39). No significant regions in the frontal lobe were noted (Baldo et al.,

2006). These findings have been replicated by other functional imaging studies (Gourovitch et

al., 2001), but additional areas of activation have been found for category fluency, namely the

left hippocampus and medial frontal cortex.

It is commonly believed that semantic fluency performance relies heavily on the integrity

of intact semantic memory networks, or the modules of long-term memory that contain

knowledge about objects, concepts, facts, as well as the meanings of words, largely localized to

the temporal structures of the language dominant hemisphere (Butters et al., 1987; Monsch et al.,

1992). Consistent with this perspective, lesion studies examining patients with temporal lobe

involvement are generally impaired on tests of semantic fluency, compared to other patient









groups and controls. Compared to healthy controls, Troster and colleagues demonstrated that

patients with left TLE generated fewer words on a semantic fluency test (supermarket fluency)

(Troster et al., 1995). This pattern of worse semantic fluency performance in LTLE compared to

controls has been replicated by others (Gleissner & Elger, 2001; N'Kaoua, 2001; Martin, Loring,

Meador, & Lee, 1990). A pattern of impaired semantic fluency performance has also been

reported in other clinical samples with temporal lobe involvement, including Alzheimer's

Disease (Randolph et al., 1993; Diaz et al., 2004). A recent meta-analysis of 995 patients (Henry

& Crawford, 2002) with a wide range of lesion etiologies found that temporal patients were more

impaired on semantic than phonemic fluency, and that those with left lateralized temporal lesions

were more impaired than those with right temporal lesions. Results on the aforementioned

neuroimaging, dual-performance, and voxel-based lesion mapping studies also confirm the

crucial role the left temporal lobe plays in retrieval of words from superordinate semantic

categories.

These findings indicate adequate performance on semantic fluency tasks is multi-

determined. It is clear that semantic fluency relies heavily upon access to the semantic memory

stores of the temporal lobe, and that damage or disease processes involving this region impairs

successful performance on semantic fluency. The role that the frontal lobes play in the controlled

search process necessary to complete the task appears to be less critical, although damage to the

frontal cortex can also impair semantic fluency to a lesser degree.

Phonemic Fluency

Phonemic fluency measures commonly consist of three trials, which require generation of

words beginning with a particular letter (Lezak, 2004). While phonemic fluency performance is

obviously dependent on the integrity of language systems as well, this measure has traditionally

been conceptualized as a measure of executive function because of the unusual demand of word









generation based on orthographic criteria. Further, the task requires the creation of nonhabitual

strategies of word retrieval based on lexical representations, and the suppression of responses

based on their meaning (Perret, 1974). Effective performance on this measure also requires

efficient organization of verbal recall, retrieval, and output, in addition to self-monitoring,

effortful self-initiation, an inhibition of previously given responses (Henry & Crawford, 2004;

Ruff et al., 1997).

Research supports the assertion that there is a relationship between the integrity of the

frontal lobes and performance on phonemic fluency. The finding of decreased phonemic fluency

in frontal-lobe patients has been reported in patients with traumatic brain injury (Jurado, et al.,

2000), left frontal and bi-frontal epilepsy (Troyer et al., 1998), dementias involving the frontal

lobes (Rosser & Hodges, 1994), and a variety of patient groups of mixed frontal-lobe pathology

(Stuss et al., 2000; Janowsky et al., 1989). In their meta-analysis, Emory and Alvarez (2006)

found that the majority of studies of frontal-lobe lesion patients reported significantly poorer

phonemic fluency scores compared to controls, although a smaller percentage found this same

difference compared to non-frontal lobe lesions patients. Another meta-analysis reported large

effect sizes (r=.52) for deficits of their frontal lobe group compared to their non-frontal group,

and deficits were largest with left frontal lesions, although patients with left focal non-frontal

lesions also showed significant impairment on phonemic fluency tests (Henry & Crawford,

2006). While the sensitivity of this measure has been demonstrated in lesion studies, its

specificity has not yet been established because a number of studies demonstrate no significant

differences between frontal patients and those with either non-frontal cortical or diffuse lesions

(Stuss et al., 1998; Miller, 1984; Pendelton et al., 1982; Perret, 1974). Further, some authors

suggest that while phonemic fluency does indeed tap an "executive" factor, the contribution of









verbal abilities to overall performance are equally important. Indeed, Ramier and Hecaen

(1970) argued that successful performance on phonemic fluency is determined by an "executive"

factor located within the frontal lobes and a "verbal" factor mediated more generally by the left

hemisphere function (presumably the language dominant hemisphere).

Neuroimaging studies of healthy controls confirm the critical role of the frontal lobes in

phonemic fluency performance, despite significant variability in task procedure and imaging

parameters. Studies have found specific areas of increased activation in the left inferior frontal

gyms (IFG), anterior cingulate (AC), and left dorsolateral prefrontal cortex (DLPFC) (Paulesu et

al., 1997; Frith et al, 1995; Frith et al., 1991). Other studies report activations in these areas

while also finding significant increases in blood/glucose to more widespread areas of the frontal

lobes (Parks et al., 1988).

In sum, evidence from lesion and neuroimaging studies suggests that phonemic fluency

relies upon the integrity of the frontal lobes, much more so than semantic fluency. Semantic

fluency on the other hand, appears to be more sensitive to the integrity of the temporal lobes and

places a larger demand on semantic memory stores. Nonetheless, contradictory reports in the

literature draw the specificity of these measures into question, as it is clear that both frontal and

temporal lobe patients can exhibit impairment on either, or both tasks. At this point in time, the

scientific literature does not provide definitive support for the notion that frontal lesions

necessarily produce disproportionate impairment on phonemic fluency, and temporal lesions

disproportionately affect semantic fluency performance. It may be that further refinements in

fluency tasks, or in ways in which fluency performance is measured, might improve the ability to

provide such a double dissociation. Providing such refinements is one purpose of the current

research.









Novel Techniques for Dissociating Frontal and Temporal Lobe Impairment

Many previous studies have attempted to differentiate frontal from nonfrontal lesions on

the basis of performance on experimental tasks designed to isolate particular aspects of frontal-

executive dysfunction. These include assessments of set shifting (McDonald et al., 2005a),

figural fluency (McDonald et al., 2005b), directed forgetting (McDonald et al., 2006), priming

(Alexander et al., 2005; Stuss et al., 1999), self-ordered pointing (Lamar & Resnick, 2004;

Petrides & Milner, 1982), source memory (Thaiss & Petrides, 2003), and structured semantic

cueing paradigms (Drane et al., 2006; Randolph et al., 1993).

This general tradition has also led to refinements of traditional oral fluency paradigms.

One such paradigm is based on the literature that posits distinct neural regions for the processing

of words that denote concrete entities, such as objects, and words that denote action or motion,

verbs.

Action/Verb Retrieval

Although a fair amount is known about the neural representation for words denoting

concrete entities ("objects") less is known about the neural basis of action word retrieval.

However, a growing body of literature across many fields, including linguistics, cognitive and

experimental psychology, neurology, and clinical psychology lends support to the idea that

distinct neural regions are involved in more highly specialized in processing information that

relates to action or movement.

Why should actions and objects have different neural bases? One theory posits that

knowledge about objects and actions is stored in association cortices adjacent to the primary

cortical regions that process these classes of stimuli (Damasio & Tranel, 1993). According to this

theory, object knowledge is stored in cortical regions adjacent to the occipito-temporal visual

stream, while action knowledge is stored adjacent to motor structures in the frontal lobe









including the prefrontal cortex, premotor cortex, and supplementary motor area (Lu et al., 2002)

that process skilled movement and action. As such, it is suspected that the frontal lobes likely

serve as a storehouse for knowledge related to movement/action, and to the extent that the major

semantic features of an object include implications for motion, a substantial portion of its' neural

network will reside within the frontal lobes. In contrast, object knowledge, based as it is on

structural representations of object form, is more dependent on the occipitotemporal visual

stream, which normally process object qualities.

Another theory posits that words are segregated based on how they were learned, thus

emphasizing the distinction between objects, which are highly visual, and actions that have

salient functional components (Warrington & Shallice, 1984). Alternatively, because actions are

captured by grammatically-rich verbs, it is possible that difficulty recognizing and producing

action words is related primarily deficits in grammatical processing (McCarthy & Warrington,

1985). Still another view holds that the deficit is largely executive in nature, and relates to the

difficulty of "mentally coordinating and manipulating the large amount of information related to

action-words" (White-Devine et al., 1996). Although the outcome of this debate remains unclear

at this point, it may be possible that several of these theories will eventually contribute to our

understanding of the mechanism underlying category-specific deficits. Further, regardless of

which theory is correct, all posit separate neural substrates for objects and actions.

The literature is beginning to draw a clear picture that category-specific deficits for action

words (verbs) do exist. The strongest support for this claim is from the lesion literature, in which

a variety of studies have demonstrated this category specific deficit. This finding was first noted

in agrammatic aphasics who demonstrated notable deficits in verb production, whereas anomic

aphasics had greater impairment in the retrieval of nouns (Miceli et al., 1984). Damasio & Tranel









(1993) present three compelling case studies; two of their patients had damage to the anterior and

middle temporal cortices and the third in the left premotor cortex. Results revealed a double

dissociation in which their first two patients had difficulty naming common nouns (pictures of

objects) whereas the third patient was unable to name actions depicted in line drawings. A

similar category-specific noun/verb effect has been shown in other groups of patients with

damage to various regions within the frontal cortex (Caramazza & Hillis, 1991; Daniele et al.,

1994; Hillis & Caramazza, 1995; Miceli et al., 1984; Rapp & Caramazza, 1998). Exploiting the

known neuropathology of various dementia types, researchers have also demonstrated action

naming impairments in groups with frontal lobe pathology, and the lack of impairment in

patients with an absence of this pathology. Although both Alzheimer's (AD) and fronto-temporal

dementia (FTD) patients displayed impairments in object and action naming compared to

controls, the discrepancy between object and action naming performance was significantly larger

for FTD than AD patients regardless of dementia severity (Cappa et al., 1998). The same group

later found that impaired action naming was not only present in FTD, but also in other patient

groups with frontal-subcortical disease, including those with supranuclear palsy and corticobasal

degeneration (Cotelli et al., 2006). Another study comparing action and object fluency in AD and

FTD confirmed Cappa's results, but further elucidated the nature of the action naming disorder.

In FTD the naming disorder was found mostly to be due to a dysexecutive deficit whereas in AD,

it was due largely to linguistic difficulties (Silveri et al., 2003). Although the possibility of a

selective verb deficit has not been explored in frontal lobe epilepsy, one study did find that verb

naming was spared in patients who had undergone LATL; a finding that is consistent with the

view that action naming is not localized to the temporal lobes (Glosser & Donofrio, 2001).









Functional imaging studies in both patients and healthy volunteers further confirm the

role of the frontal lobes in the retrieval of action words. In fact, one of the earliest tasks used in

PET and fMRI studies was a task in which participants were shown a picture of an object

("ball") and asked to generate a verb ("kick") for the object. These early imaging studies found

that the left inferior frontal gyms (IFG) was activated during these tasks (Petersen et al., 1989;

Petersen, Fox, Snyder & Raichle, 1990; Raichle et al., 1994). More recent studies have

confirmed the role of the frontal cortices in naming and generating action words. Thomspon-

Schill found that patients with damage to the left IFG not only had more difficulty generating

semantically appropriate verbs, but also made more errors on their task than did patient or elderly

control groups (Thompson-Schill, 1998). Other fMRI findings using varying task demands found

similar activations in the left inferior prefrontal cortex (Perani et al., 1999; Shapiro, Moo &

Caramazza, 2006; Tyler et al, 2004). One recent PET study found that naming actions was

correlated with increased glucose utilization in the left frontal operculum, left posterior middle

frontal gyms, and left and right parietal lobule (Damasio et al., 2001).

Until recently, the assessment of verb retrieval abilities has been limited to action naming

paradigms, similar to an "action" analog of traditional naming tests, such as the Boston Naming

Test (BNT). While these paradigms are useful in assessing verb naming impairments, they all

require identification of a verb associated with a graphically depicted image (Obler & Albert,

1979) as opposed to free generation of action-related words. Over the past several years,

however, a small body of literature using action fluency paradigms has emerged. Action fluency

paradigms assess the spontaneous production of verbs, with the instructions "tell me as many

different things as you can think of that people do" (Piatt, Fields, Paolo, & Troster, 1999).









Preliminary evidence from action fluency studies suggests that performance on this task

is indeed sensitive to frontal lobe dysfunction. Patients with known frontal-subcortical damage

were significantly impaired on action fluency when compared to healthy controls. Patients with

frontal-subcortical pathology secondary to HIV-1 infection performed similarly to healthy

controls on measures of semantic fluency, but were significantly impaired on the action fluency

measure (Woods et al., 2005). When Parkinson's patients with (PDD) and without dementia

(PDND) and healthy controls were compared on action, phonemic, and semantic fluency tasks,

PDD patients performed worse on all three measures. However, performance on the action

fluency task was differentially more difficult for the PDD group than semantic or phonemic

fluency, relative to the control and non-dementia groups (Piatt et al., 1999b). The authors

conclude that the measure was both sensitive and specific to frontal-subcortical disease.

Three studies have examined the construct validity of the action fluency test as a measure

of frontal lobe functioning and/or executive function. Piatt Fields, Paolo & Troster (1999)

demonstrated the convergent validity in a sample of healthy older adults. They found that action

fluency performance was significantly related to several measures of executive function (i.e.

TMT-B, WCST) but not with common measures of temporal lobe functioning (i.e. BNT, Logical

Memory from the Wechsler Memory Scale). While action fluency shared common variance with

other measures of executive function, the test also seemed to measure a component of executive

functioning not tapped by more traditional tasks. Woods et al (2005) found similar relationships

with putative measures of executive function in healthy young volunteers, but found no

relationship with measures traditionally associated with the posterior cortex. In sum, there is

support for dissociation between action and object naming, with the naming of actions being

dependent on anterior brain structures, namely the frontal lobes. Further, preliminary findings









indicate that action fluency is a novel paradigm that may not only be sensitive, but also specific

to frontal lobe functioning. To our knowledge, this paradigm has been applied only to patients

with HIV and Parkinson's Disease and normal older and younger controls, and has not been used

in patients with focal epilepsy.

Proper Name Retrieval

The difficulty in isolating patients with frontal lobe damage lies not only in identifying

tasks sensitive to frontal lobe dysfunction, but also in designing equivalent tasks sensitive only to

temporal lobe damage in order to doubly dissociate performance on neuropsychological tests.

While the mesial temporal lobe is critically important in episodic memory, the lateral (cortical)

aspects of the temporal lobes are likely critical in the storage and maintenance of semantic

memory, or knowledge of objects, facts, and names. Patients with damage to the anterior and

lateral portions of the temporal lobes often have difficulty on tasks tapping semantic memory

stores. This is particularly true in the case of proper names. The specific difficulty in producing

proper names has generally been attributed to their semantic uniqueness, or the fact that these

names refer to unique entities (Semenza & Zettin, 1989; Grabowski et al., 2001). While common

names refer to concepts, or a set of attributes that are shared by multiple entities within the same

concept, proper names do not inherently contain attributes in and of themselves and are merely

expressions by which we refer to an individual person or item. Because of this, it is thought that

widespread neural networks support the representation of common nouns, while proper nouns

are thought to hold rather fragile "associations" with their unique reference (Martins & Farraj ota,

2007). In addition, difficulty in retrieval of common nouns is often abated by the fact that they

can often be substituted with synonyms, whereas this is not usually possible with proper names

(Bredart, 1993).









Regarding the neural representation of proper names, several cases studies have shown

that lesions to the left anterior temporal and temporal polar regions selectively disrupt the

retrieval and production of proper names (Damasio et al., 1996; Harris & Kay, 1995; McKenna

& Warrington, 1980). Two lesion studies demonstrated this effect across naming paradigms

(naming to pictures, naming to description), and name generation tasks (actors, sports figures).

The dissociation of common and proper name impairments was also demonstrated in a stroke

patient (ACB) who suffered an ischemic lesion that involved that temporal neocortex and

temporal pole (Martins & Farrajota, 2007). Subsequently, he was unable to recall proper names

(particularly those that referred to well-known figures, such as politicians) while his ability to

produce common names of objects was relatively spared. Similar deficits in proper naming

abilities were documented in a patient who underwent left ATL surgery for refractory epilepsy

(Fukatsu, et al., 1999). This patient was able to accurately perceive pictures of faces, but was

able to name only 25% of his acquaintances from photos, and even fewer from verbal description

(24%). He named only 4 out of 25 famous faces correctly. He was able to name 90/100 pictures

of common items, however (animals, furniture, tools, insects). Further, his fluency for common

names was double his fluency for proper names. Similar, though less dramatic findings of proper

name impairment have been reported in patients who underwent LATL for epilepsy relief (Barr,

Goldberg, Wasserstein & Novelly, 1990; Tsukiura et al., 2002; Seidenberg et al., 2002; Glosser,

Salvucci, Chiaravalloti, 2003). In these studies, many of the LATL patients had impaired naming

of famous faces compared to controls, but were able to recognize or subsequently provide

semantic information about them, apart from their actual name.

Because the majority of these studies involve the naming of persons in response to visual

representation, it raises the question of whether this impairment is related to the specificity of the









task (i.e. retrieving unique names) or to preferential processing of facial stimuli by the anterior

temporal lobe. Several neuroimaging studies have tried to address this question. Regardless of

presentation format (printed names versus photo) or presentation modality (visual versus

auditory), similar areas of activation are noted (Tempini et al., 1998; Tranel, Grabowski, Lyon &

Damasio, 2005). In addition, this question has been addressed directly by comparing retrieval of

names of other unique entities (landmarks and buildings) to retrieval of proper names of persons.

The results of these studies (Milders, 2000; Tranel, 2006; Grabowski et al., 2001) confirm the

hypothesis that portions of the left temporal lobe (anterior temporal lobe, temporal pole) are in

fact specialized for the retrieval of unique entities as a whole, not only entities that contain

human features.

Apart from a limited number of lesion studies that have used the design experimentally,

the assessment of unique (or proper) naming abilities has been limited to paradigms that require

identification of a name associated with pictures or photographs. Although there is sufficient

evidence to support the idea that the retrieval of proper names depends on the integrity of

anterior portions of the temporal lobe, and taps a unique aspect of semantic processes, these

hypotheses have not been directly tested in this format in clinical populations, and the ability to

generate proper names without corresponding visual stimuli has not been examined.

Qualitative Analysis

Tests of verbal fluency, regardless of type, are generally scored by tallying the total

number of words generated, minus errors or repetitions. While this score is an accurate

measurement of fluency output, it provides minimal information about how the task is

completed, and as previously discussed, may be limited in accurately characterizing

performances by different groups of patients because the same score can be obtained in

qualitatively different ways. Because fluency score is likely multi-determined, and affected by









difficulty initiating or maintaining performance, faulty search and retrieval strategies, degraded

semantic memory stores, failure to maintain set, and self-monitoring failures, a more thorough

analysis of the cognitive processes involved in task performance is warranted.

Aspects of fluency performance, including perseverations and intrusions, have been

assessed in a variety of clinical and healthy populations (Reverberi, Laiacona, & Capitani, 2006;

Warrington, 2000; Martin & Fedio, 1983; Troster et al., 1989) and recently, Troyer, Moscovitch,

& Wincour (1997) developed a methodology for examining organizational retrieval processes

involved in word generation. They suggest that optimal fluency performance is composed of the

production of semantic (i.e. apples, bananas, grapes) or phonemically-related (i.e. far, fat, fast)

"clusters" of words, and when one cluster is exhausted, a switch is made to another cluster. As

such, they envision two important aspects of fluency performance; clustering, which is the

production appropriate words within the subcategories, and switching, the ability to shift

between said subcategories. Clustering is thought to rely heavily upon organized access of

semantic memory stores (more strongly localized in the temporal lobe), while switching is

thought to rely more heavily upon cognitive flexibility, ability to shift set, disengagement from

previous responses, and strategic search (more strongly localized in the frontal lobe; Troyer &

Moscovitch, 2006). Some evidence suggests that while clustering relies on relatively automatic

cognitive process, switching is thought to involve effortful processing (Rende, Ramsberger, &

Miyake, 2002).

Data from healthy young and old volunteers has suggested that clustering and switching

are indeed dissociable processes. Clustering and switching scores were both related to total

fluency score on semantic fluency measures, but the switching score was more uniquely related

to overall phonemic fluency, consistent with the view that both rely more heavily on frontal-









executive processes (Troyer, Moscovitch & Wincour, 1997). The finding that divided-attention

tasks (concurrent finger-tapping) disrupted switching but not clustering also supports this

assertion.

Given that these processes have been identified as dissociable in healthy volunteers and

have shown differential relationships with overall fluency scores, others examined whether they

would be differentially affected by neurological disorders. As predicted, patients with temporal

lobe epilepsy showed decreased clustering on semantic fluency tests, while patients with frontal

lobe epilepsy exhibited decreased switches on both phonemic and semantic fluency tests. Troyer

et al. (1998a) determined that the best indices for discriminating these patients were phonemic-

switching and semantic-clustering scores. Early AD patients, who have known temporal lobe

pathology, showed reduced cluster size on both types of fluency (Troyer et al., 1998b). Patients

with frontal and/or subcortical disease demonstrated the expected pattern of relatively intact

semantic cluster size, but decreased switching (Demakis et al., 2003; Ho et al., 2002; Troster et

al., 1998) as did patients with psychiatric disease known to affect frontal-lobe functioning

(Fossati et al., 2003; Robert et al., 1998).

Quantitative fluency scores are certainly sensitive to frank pathologies such as those

underlying Alzheimer's Disease and aphasia, and can be also be sensitive to milder forms of

pathology in some cases. However, impairment of this score can be due to heterogeneous causes,

which greatly limits or precludes interpretation about the underlying cognitive processes

responsible for the deficit. A qualitative analysis of fluency performance, however, helps to

elucidate the mechanisms by which the task is completed and may shed light on the nature of

fluency impairment. This qualitative analysis appears to be most helpful in identifying

components of fluency performance due to frontal and temporal lobe impairment.









Summary

Elucidating a pattern of neuropsychological test performance characteristic of frontal and

temporal lobe epilepsy is complicated at best, but remains important for a number of reasons.

First, identification of patterned impairments allows for better and more accurate identification

of seizure onset, which is essential in treatment planning. Beyond the clinical and practical

importance, a more thorough understanding of these cognitive impairments further extends our

scientific knowledge about the neural substrate of these cognitive processes. Presently, these

cognitive patterns are poorly defined and overlapping, partly due to the lack of specificity of

many of our tests.

In particular, many of our tests thought to identify frontal lobe dysfunction appear

sensitive, but not specific. These include tests thought to assess mental flexibility, set-shifting,

response inhibition, and working memory. Traditional measures of fluency are commonly used

to identify patterns of performance of frontal and temporal lobe epilepsy patients, however, both

patient groups may exhibit deficits on both types of tests, affording them minimal discriminative

validity.

Despite the relative lack of specific standardized assessments, the scientific literature

provides insight into the type of impairments that may exist with damage to either the frontal or

temporal cortices. Specifically, damage to the language dominant temporal lobe produces

deficits in the retrieval or names of unique entities (people or places), whereas damage to the

frontal cortex, particularly left lateralized damage, produces impairments in the retrieval of

words denoting actions. In conjunction with the test content, a qualitative examination of the

cognitive strategy involved may also useful predictive value.

The current study seeks to incorporate this material-specific content into traditional test

paradigms in order to further explore the unique cognitive deficits associated with localized









neural dysfunction. By manipulating the retrieval demands involved and examining the cognitive

strategies employed, we hope to more accurately discriminate between patient groups and to

advance our understanding about the neural specificity of these brain regions. Specific aims of

the study are listed below.









CHAPTER 2
SPECIFIC AIMS & HYPOTHESES OF CURRENT STUDY

Aim 1

The first aim of the study was to characterize performance of patients with either frontal or

left temporal lobe epilepsy, and matched healthy controls on a panel of verbal fluency tests, that

includes both traditional measures of semantic and phonemic fluency, and experimental measures

of action and proper name fluency. We hypothesized that overall fluency score on action and

proper noun fluency would doubly dissociate patients with frontal and temporal lobe epilepsy,

with frontal lobe patients performing worse on action fluency and temporal lobe patients

exhibiting comparative deficits on tests of proper name fluency. We also suspected that patients

with temporal lobe epilepsy would also evidence impaired semantic fluency (but not phonemic

fluency), while patients with frontal lobe would demonstrate impairments on phonemic fluency

(but not semantic fluency). All patient groups were predicted to generate fewer total words than

controls due to the overall effect of their neurological disorder.

Aim 2

The second aim of the study was to compare the clinical utility and predictive validity of

experimental versus traditional fluency measures in identifying seizure location and

lateralization, and in discriminating between temporal and frontal groups. We predicted that the

experimental fluency measures would have similar, and possibly more favorable operating

characteristics (sensitivity, specificity, positive and negative predictive value) than traditional

measures of fluency, and that our panel of fluency measures would be effective in accurately

discriminating frontal and temporal lobe patients.









Aim 3

The third aim was to establish the psychometric properties of the experimental fluency

measures (convergent and discriminant validity) using traditional measures of frontal and

temporal lobe dysfunction and to compare the predictive power of various fluency tests with

more traditional neuropsychological measures. We hypothesized that performance on tests of

common and proper noun fluency would be related to measures of language and

semantic/episodic memory (Wechsler Memory Scale-Logical Memory, Boston Naming Test),

while action and phonemic fluency scores will exhibit moderate relationships with traditional

measures of executive function (Wisconsin Card Sorting Test, Trail Making Test B). However,

as these tests tap varied aspects of cognitive functioning whose neural instantiations exist within

the frontal lobe, we thought that this fluency measure may comprise a new dimension of

executive function not assessed by other measures.

Aim 4

The fourth aim was to determine whether a qualitative analysis of fluency performance (i.e.,

clustering and switching performance) would dissociate performance of patients with FLE, TLE,

and healthy controls. Consistent with previous lesion studies, we expected that patients with TLE

would exhibit an average number of switches, but reduced cluster size, particularly on tests that

rely more heavily on semantic memory (proper name fluency, common noun fluency).

Conversely, we thought that patients with FLE would display the reverse pattern of spared

semantic cluster size, but reduced number of switches, predominantly on tests of phonemic and

action fluency.









CHAPTER 3
SUBJECTS & METHODS

Study Participants

Participants in this study included patients with documented epilepsy localized either to

the frontal or temporal lobes. A healthy control group with no current or past history of

neurological disease was also recruited for participation in the study. Initial power analyses,

computed from data provided in Troyer et al. (1997), indicated that with an evenly distributed

sample of 30 (10 patients with left frontal lobe epilepsy, 10 patients with left temporal lobe

epilepsy, and 10 healthy controls), our study would be powered adequately (Critical F= 3.55,

Actual power = .975, a error probability = .05) to detect overall group differences in fluency

score, our primary aim of the study. However, additional analyses computed from data in Troyer

et al. (1997) also suggested that as few as six participants were needed per group (Critical F=

3.88, Actual power = .955, a error probability = .05).

Initial investigation of pre-surgical patient flow over the past several years suggested our

patient recruitment goals were feasible in a pre-surgical population over a ten to twelve month

recruitment period. While the initial goal of the study was to recruit pre-surgical patients with

either language-dominant temporal lobe epilepsy (i.e., left-hemisphere) or patients with left

frontal epilepsy, this goal could not be achieved even with extended recruitment over a period of

eighteen months. In eighteen months, we were successful at recruiting only five presurgical

patients with left temporal lobe epilepsy who met all inclusion/exclusion criteria and four

patients with frontal lobe epilepsy (one right, two left, one bifrontal).

Because of the significant difficulty recruiting patients pre-surgically, post-surgical

patient data was collected concurrently. Combining both pre-surgical and post-surgical patients,

we were able to collect data on fourteen patients with left temporal lobe epilepsy, seven patients









with frontal lobe epilepsy, as well as eighteen patients with right temporal lobe epilepsy. By

combining both pre-and post-surgical data, we were able to meet our sample size requirements in

our left temporal lobe group. However, even when combining pre- and post-surgical data we

were not able to obtain ten patients with left frontal epilepsy, which was our initial goal. Because

of this, we combined patients with left frontal, right frontal, or bifrontal epilepsy to compose our

"frontal lobe" group, for a total of seven frontal lobe epilepsy patients (Table 3-1).

Participants included in the final study were fourteen patients (pre- and post-surgical)

who had epilepsy localized by EEG to the left temporal lobe, seven patients (pre- and post-

surgical) with EEG documented epilepsy of the frontal lobes (left, right, or bilateral) and twenty

healthy controls. Lateralization and localization of seizure foci was determined by consensus

diagnosis using data from Phase I EEG monitoring, MRI, patient history, and cognitive test

results. At the present time, six of the nine pre-surgical patients have proceeded to surgery.

Differences in demographic and clinical characteristics were assessed using one-way analysis of

variance (ANOVA) and chi-square tests. Patients and control groups were well matched for age

(F (2, 40) = .08, p>.05), education (F (2, 40) = 3.1, p>.05), gender (X2 (2) = .03, p>.05),

handedness (X2 (2) = 4.09, p>.05), race (X2 (2) = 3.51, p>.05), and WASI full scale IQ (F (2,40)

= 2.62, p >.05) (Table 3-2). On average, our patient and control groups were around forty years

of age, had slightly more than a high school education, and were composed largely of

Caucasians. While controls had slightly higher full-scale IQ's, patients and controls did not differ

significantly and overall intellectual functioning for all three groups was in the average range.

The groups were predominantly right handed (X2 (2) = 4.09, p>.05), and were composed of

slightly more women than men. Patient demographics are summarized in Tables 3-1 and 3-2.









Patient groups did not differ significantly on language dominance (t (20) = .80,p>.05) or

age of seizure onset (t (20) = .28, p >.05). In general, temporal lobe and frontal lobe patients first

developed seizures during their teenage years, though there was a wide range in the age of

seizure onset amongst the patients (LTL M=16.9, SD=15.2; FL M=13.5, SD=11.7) (Table 3-2).

Four of our TL patients and two of our FL patients had a history of psychiatric illness or

diagnosis (X2 (1) = .04, p>.05). With regard to risk factors for epilepsy, thirty-five and twenty-

eight percent of TL and FL patients had a family history of epilepsy, respectively (X2 (1) = .10,

p>.05). Four TL patients had suffered a mild-to-moderate traumatic brain injury (TBI) and three

of the FL patients experienced a mild-to-moderate TBI (X2 (1) = .2.14, p>.05). Slightly more TL

patients had a history of child illness, though this was not statistically significant (X2 (1) = 3.28,

p=.07). In general, the patients in this study did not have a history of febrile seizures (X2 (1) =

2.14, p>.05). Results of magnetic resonance imaging scans (MRI) revealed that ten of the TL

patients had lesions consistent with mesial temporal sclerosis (MTS), and four of our FL patients

had lesions or other neural anomalies (X2 (1) = 5.57,p =.06).

Epilepsy Patients

Pre-surgical patients

All pre-surgical patients selected for inclusion in this study had medication refractory

epilepsy, documented by a board-certified neurologist at the University of Florida

Comprehensive Epilepsy Program (UFCEP), and had experienced uncontrolled seizures under at

least two medication regimens for at least an 18-month period. Participants were identified and

recruited from the Neurology and Psychology Clinics at Shands Hospital and were screened to

determine if they met inclusion criteria for the study, as approved by the Institutional Review

Board (IRB) at the University of Florida. Inclusion criteria for all patients were: 1) 18 years of

age or older, 2) documented epilepsy of either the temporal or frontal lobes, and 3) English as a









first language. Exclusion criteria were: 1) history of severe developmental disorder, or mental

retardation resulting in IQ < 69, 2) history of Axis I psychiatric disturbance that resulted in

inpatient hospitalization, 3) history of substance abuse/dependence, using DSM-IV criteria, 4)

first language other than English, 5) lesional temporal-lobe epilepsy, or 6) presence of

neurological disease in addition to epilepsy (e.g., radiation or chemotherapy for brain cancer in

the past year; head trauma resulting in moderate to severe brain injury).

Patients were approached during either their Phase I (inpatient) video-EEG hospital stay

or during their outpatient appointment in Neuropsychology, were presented with information

about the study, and were asked if they would like to participate. Patients who agreed to

participate gave informed consent and were tested in their hospital room as an inpatient in

Shands hospital, or arranged testing on another day as an outpatient. In a few instances, testing

could not be completed in one session and the patient was seen twice. Relevant demographic and

medical information was also gathered from the participant's medical record. Relevant

demographic information collected included age, gender, educational and occupational

attainment, reported handedness, and ethnicity. Medical and seizure related variables were also

recorded, and included age of first seizure, current medications regimen, seizure frequency and

duration, and Wada memory and language dominance.

Over the course of eighteen months, approximately 150 pre-surgical patients were

screened for the study via medical record review. Of the patients screened, we recruited 37

patients for the study, and of those patients, nine met full inclusion criteria for the present study;

five were determined to have clearly lateralized left temporal lobe seizures and four were

determined to have frontal lobe seizures (one right, two left, one bilateral). The remaining

patients were determined to have non-epileptic seizures, seizures originating from another foci,









had right or mixed language dominance, were unable to be classified based on EEG recordings,

or did not meet inclusion/exclusion criteria. Classification of seizure onset was determined by a

board-certified neurologist who reviewed EEG recordings, clinical notes, and patient history. A

summary of pre-surgical patient characteristics is included in Table 3-1 and Table 3-2.

Post-surgical patients

Post-surgical epilepsy patients were patients who had undergone surgery between 2000

and 2007, and had undergone either an anterior temporal lobectomy or a frontal cortical

resection. These patients were recruited from the Departments of Neurosurgery and

Neuropsychology at the University of Florida. Patients were identified through clinical databases

and were selected for recruitment based on surgery type, medical record review, and date of

surgery. Patients were recruited throughout the entire state of Florida.

Eligible patients were contacted via letter and provided with information about the

purpose of the study, and its potential benefits and risks. They were provided with a self-

addressed stamped postcard so they could indicate a desire to participate in the study, decline

participation, or inquire further about study details. Patients who returned the postcard and were

interested in participating or wanted more information were then contacted and given the

opportunity to inquire further or schedule an appointment for cognitive testing. They were also

screened briefly over the phone, after which the cognitive testing was arranged. Testing took

place either at Shands Hospital, or if the patient was unable to travel to Gainesville, at the

patient's home.

We screened approximately 1,000 patients who had undergone epilepsy-related

neurosurgery (i.e., grids, resections, vagus nerve stimulator placement) between the years 2000

and 2007, and attempted to recruit approximately 125 by letter. The 125 persons contacted met









criteria for study enrollment and had current mailing addresses. Of the 125 contacted,

approximately 50 responded with an interest in the study. Of the 50 who expressed interest, 30

were enrolled in the study. The remaining 20 patients either declined participation, were unable

to be scheduled due to distance (> 5 hours away), or had travel, family, or work conflicts.

Patients who participated in the study included nine patients with left temporal lobe resections,

eighteen patients with right temporal lobe resections, and three patients who had undergone right

frontal cortical resections. A summary of post-surgical patient characteristics is presented in

Tables 3-1 and 3-2.

Healthy controls

We attempted to recruit family members of patients to serve as healthy volunteer (i.e.

control) participants. When patients were recruited into the study, family members who were

present were also informed about the study and given the option to be screened and participate as

a member of the healthy control group. Post-surgical patients were informed on the phone about

the option of family members participating. These individuals were often tested the same day as

their family member, in a quiet room in Shands hospital, although several were tested on

alternative dates. The remainder of our healthy controls were recruited in Gainesville, Florida

through the use of fliers. When interested persons called to inquire about the study, they were

informed of its purpose and procedures. They were also given a brief screening measure over the

phone, consistent with IRB protocol, to determine if they were eligible to participate. Volunteers

who were eligible and interested in participating engaged in a single testing session in a quiet

testing room on the ground floor of Shands Hospital. During this session, volunteers first gave

informed consent in person, and displayed an understanding of the testing procedures.









We attempted to match this group on age and education variables with our patient groups.

Exclusion criteria for our controls were: 1) Age younger than 18, 2) history of severe

developmental disability or mental retardation resulting in IQ < 69, 3) history of significant Axis

I psychiatric disturbance that resulted in inpatient hospitalization, 4) history of substance abuse

or dependence, 5) neurological illness (e.g., epilepsy, cerebrovascular disease, brain tumor, or

head trauma resulting moderate to severe head injury), 6) current radiation or chemotherapy

treatment (within 1 year), or 7) English as a second language.

We recruited twenty healthy controls to serve as volunteers in our study, eight of whom

were family members of patients. A summary of control characteristics is included in Table 3-2.

Measures

Pre-surgical Patients

Typically, pre-surgical epilepsy patients are routinely administered a standard

neuropsychological test battery (SNB) as part of their Phase I evaluation through the UFCEP.

Most patients in our study were administered the SNB, though some were still in the process of

their pre-surgical evaluation, and had not yet undergone clinical neuropsychological testing. As

part of the SNB evaluation, subjects were administered tests of verbal and non-verbal memory,

naming, fluency, "executive function", visuoconstructional and visuospatial ability, attention,

mood, and intellectual functioning. A list of measures traditionally administered is listed in

Appendix A. Because many measures of interest are routinely administered to patients, and

because re-testing patients on the same tests may have produced biased results, data for selected

tests from the SNB was used in the current study. Measures of particular interest from the SNB

included assessments of verbal memory (Logical Memory Stories from the Wechsler Memory

Test-III [WMS-III]), language (Boston Naming Test [BNT]), and measures of"executive









function" (Wisconsin Card Sorting Test [WCST], Trail Making Test B [TMT B], and WAIS-III

Digit Span. As a detailed description of many of these tests has been provided elsewhere (Lezak

et al., 2004; Spreen & Strauss, 2006) and are reviewed briefly in Appendix A, they will not be

described in further detail here.

The study measures were administered during a separate experimental testing session.

These measures included a standard assessment of phonemic fluency, using three 60-second

trials of word generation beginning with the letters "C", "F", and "L". These phonemic probes

were chosen to be distinct from the phonemic fluency trial (FAS) given during the SNB.

Analyses were computed using the letter "C" only so that overall score would be comparable to

that of other fluency measures (i.e., a single trial) although analyses with overall phonemic

fluency score (i.e., C+F+L) were computed as well. The semantic fluency category (i.e.

Supermarket items) was also chosen for its non-overlap with category, "Animals", used in the

SNB. As a hierarchically organized "supercategory", supermarket items also permit an analysis

of how patients' fluency performance reflects retrieval from multiple semantic subcategories

within a trial. Total score for this was also the number of correct words generated within 60

seconds. Clustering and switching was also evaluated, as described below.

Two experimental measures of fluency were administered. These measures are similar in

administration time and scoring format to the previous measures, but differ in content. For the

Action Fluency Test, participants were instructed to "List as many different things that [you] can

think of that people can do". The total score for this task is the number of correct target words,

minus words denoting actions not performed by people (e.g., "molt" or "photosynthesize").

Questionable answers (homonyms, i.e. bear-bare), or words with ambiguous grammatical roles

(e.g., table) were queried by the examiner to determine if the word was an intrusion, and was









scored accordingly. For the Famous Name Fluency Test, patients were be instructed to "List as

many different names of famous or well-known people that you can think of', within a minute.

Total score for this measure represented the number of correct names generated, minus names

that could not be verified as famous persons, or intrusions or repetitions. As the format for these

tests is similar, the administration of these four trials were counterbalanced across participants to

account for any practice effect related to familiarity with test administration procedures (i.e.

ABCD, DCBA, BDAC, CADB).

Qualitative analysis of the fluency protocols was completed in accordance with procedures

described elsewhere (Troyer & Moscovitch, 2006). Briefly, clusters on phonemic fluency were

defined as "successively generated words that begin with the first two letters, differ only by a

vowel sound, rhyme, or are homonyms". Semantic clusters were successively generated words

that belong to a semantic subcategory, such as fruits, vegetables, meat, dried goods, and dairy, on

supermarket fluency. Pilot data for our study revealed several naturally occurring clusters for

action and name fluency. Clusters that emerged for action fluency were actions with the hands,

feet/legs, facial gestures, grooming, household actions, work actions, emotional actions, actions

involving language, and actions involving rest/relaxation. Clusters on famous name fluency were

actors/actresses, politicians, sports figures, TV personalities, and musicians/singers.

Additionally, on this test clusters could be defined temporally (TV personalities from the 1980's)

or by relationships (either personal or professional). As is customary, the cluster size was

counted beginning with the second word within each cluster. See Appendix B for examples.

Switching scores were defined as transitions between clusters. Although some controversy

exists around this scoring procedure, the raw number of switches is used as the score, rather than

a score corrected for total words produced. Because switching is in part, responsible for overall









score, "adjusting switches according to the total words generated would be akin to correcting a

cause for its effect" (Troyer & Moscovitch, 2006). To further emphasize this point, only raw

switching score has been show to produce meaningful data (Troyer et al., 1998b). Errors and

intrusions were also included in scoring clusters and switches, as they are thought to provide

useful information about the underlying cognitive strategy, but are always excluded from total

overall fluency score. All fluency measures were double-scored to ensure accuracy in calculation

of total score, errors, clusters, and switches.

Finally, apart from measures of cognitive functioning, all study participants were

administered a brief questionnaire assessing demographic variables and the Edinburgh

Handedness Inventory (Oldfield, 1971).

Additional Measures

In addition to the SNB and experimental measures listed above, two additional measures

were given to study participants, The Action Naming Test (Obler & Albert, 1979), and a

modified version of the Famous Faces Naming Test (Seidenberg et al., 2002). While these

measures were not related directly to the primary aims of the study, they were administered to

study participants for several reasons.

First, the experimental fluency paradigms proposed for this study were based in part on the

naming literature which provides evidence for the category-specific deficits for actions and

proper nouns/names. This body of naming literature posits distinct neural substrates for action

naming (i.e. the frontal cortices) and person naming (i.e. the anterior temporal cortices of the left

hemisphere). While there is similar (but preliminary) support for a comparable substrate for

action fluency, this paradigm has not been used extensively in the literature, and has not been

used in epilepsy populations. Further, the name fluency paradigm has not been used widely

either in clinical or research arenas, apart from its use in an isolated number of case studies.









While we hypothesized these deficits would be apparent in a fluency paradigm, this extension

has not yet been documented empirically. As such, the administration of naming analogues to

our fluency paradigms allowed us to examine the pattern of deficits across naming and fluency

tests in order to determine whether these unique category-specific deficits were isolated to

naming, or are in fact, present in a fluency format as well.

Further, to our knowledge, the discrepancy between action and proper noun naming has

not been explored in clinical epilepsy. The addition of these tests to our battery not only allowed

for comparisons between naming and fluency paradigms, but also provided interesting

information about various category-specific naming deficits which have been relatively

unexplored in this population. Additionally, these naming tests were examined in conjunction

with performance on a common object naming test, the BNT. A brief summary of these tests

follows:

Action Naming: The Action Naming Test (ANT; Obler & Albert, 1979) contains a series

of 55 black-and-white line drawings and was modeled after the Boston Naming Test (BNT). The

items in this test, however, are a series of line drawings depicting actions. Sample actions include

running, swimming, reading, curtsying, and exercising. Test items range in difficulty in

ascending order. Study participants were shown one picture at a time and were instructed to "tell

what is happening in the picture", preferably with one word responses. Similarly to the BNT,

phonemic cues were given if the participant could not name the action, although successful

naming of these items are scored as incorrect. Total score was the number of correct answers.

Famous Face Naming: The Famous Face Naming Test (Seidenberg et al., 2002) is a test

that contains 100 black-and-white photographs of famous or well-known persons. A shortened

version, containing 48 stimuli, was used for this study. The sub-set of stimuli was selected in









order to shorten test administration time, and make it comparable in length to our other naming

tests. Stimuli were excluded from the study if they contained features that aided in identification

(i.e. shirts with sports logos, war uniforms) or if the picture quality was deemed unsatisfactory.

The stimuli are head-shots of famous persons sampled across decades since the 1960's, and

contain images of athletes, presidents, actors, singers, politicians, and other newsworthy

individuals. Sample faces include Kobe Bryant, Peter Jennings, Bob Barker, and Sting.

Participants were initially asked if they recognized the person pictured and asked to provide any

descriptive information about them (i.e., a comedian; the boxer who bit someone's ear off; the

guy who hosts the New Year's Eve celebration in Times Square). They were then instructed to

provide the name of the person in the picture if able. No cues were given. Because recognition of

test items varied across participants, the total score was corrected for the number of famous faces

recognized (i.e., correct names produced/recognized multiplied by 100; Drane et al., 2008).

Post-surgical Patients and Healthy Controls

Healthy volunteers and post-surgical patients were administered identical measures. This

included four fluency tests (phonemic, semantic, action, and proper name), and two naming tests

(action and famous faces). Because healthy controls and post-surgical patients had not been

administered the SNB, additional testing was completed in order to obtain scores on measures of

language, verbal memory, and executive function. Accordingly, they were administered the

WASI, WMS-III LM, WCST, TMT-B, Digit Span from the WAIS-III, and BNT.

Pre-surgical, post-surgical and control participants were all compensated $10.00 per hour

of time and given a $3.00 parking voucher. Average time spent in completion of this study was

an hour for pre-surgical patients and two-and-a-half hours for post-surgical and control

participants.









Table 3-1. Patient seizure characteristics
EEG Interpretation/Resection Site Surgery?
Temporal
Pre left temporal lobe and NES No
Pre bilateral (left > right) temporal Yes
Pre left temporal lobe Yes
Pre predominantly left temporal; 1 bilateral onset Yes
Pre left temporal lobe Yes
Post LATL 2001
Post LATL 2006
Post LATL 2004
Post LATL 2003
Post LATL 2002
Post LATL 2003
Post LATL 2006
Post LATL 2003
Post LATL 2005

Frontal
Pre left frontal lobe Yes
Pre right frontal lobe Yes
Pre bilateral frontal (left > right) No
Pre left frontal lobe No
Post Right Frontal Cortical Resection 2007
Post Right Frontal Cortical Resection 2006
Post Right Frontal Cortical Resection 2005









Table 3-2. Demographic and clinical characteristics
LTL FL


Age (years) 41.1 (13.2) 39.4 (12.9) 4
Education (years) 13.5 (2.6) 12.7 (2.5) 1
Gender (M/F) 6/8 3/4
Handedness (R/L) 12/1 6/1
Race (Cau./AA/His.) 12/1 7/0/0
Full-Scale IQ 96.1 (12.3) 94.0(14.6) 1
Language Dom. (L/R) 14/0 5/1
Age of Seizure Onset 16.9 (15.2) 13.5 (11.7)
Note: Means are presented, standard deviations in parenthesis.


Controls
20
0.9(15.7)
15.3 (2.7)
8/12
11/1
16/3/1
04.1 (9.8)









CHAPTER 4
RESULTS

The results of our analyses are presented below as they relate to the specific aims of the

study:

Aim 1

The first aim of our study was to characterize performance of patients with either frontal or

temporal lobe epilepsy, and matched healthy controls on a panel of verbal fluency measures,

including traditional measures of semantic and phonemic fluency, and experimental measures

including action and proper name fluency.

Because we were interested in determining whether a panel of fluency tests incorporating

traditional and experimental measures would detect group differences in our patient groups, we

first computed separate multivariate analysis of variance tests (MANOVA's) for the tests we

hypothesized would be sensitive to TL pathology (semantic and name fluency tests) and the tests

hypothesized to be sensitive to FL pathology (action and phonemic fluency tests). Using Pillai's

trace, there was not an overall group difference for our patients in fluency performance on

phonemic and action tests (V=.02, F (2, 18) = .135, p>.05, 12=.015). There was, however, an

overall significant group difference for the semantic and name fluency panel (V=.28, F (2, 18) =

3.54,p=.05, r2=.28). Results of the omnibus MANOVA's are presented in Table 4-2. Univariate

test results are presented below.

When comparing patients and controls, healthy controls outperformed patients on all

measures of fluency including standard measures of semantic (supermarket) and phonemic (letter

fluency) fluency, as well as experimental measures of fluency, including action and famous

name tests. Overall univariate tests of analysis of variance (ANOVA's) revealed significant main

group effects for action (F(2, 38) =7.90 p=.001, 12=.54) and name fluency (F(2, 38) = 17.02









p=.000, 12=.68) tests, and there were trends for significance for the semantic (F(2, 38) = 2.94,

p=.065, r2=.36) and phonemic fluency (F(2, 38) = 2.73,p=.07, r12=.35) measures. Planned post-

hoc contrasts revealed significant group differences between controls and both patient groups on

action fluency (LTL < Control, p<.005; FL< Control, p<.05) and name fluency (LTL< Control,

p<.0001; FL< Control, p<.01). When combining scores across phonemic fluency trials ("C" plus

"F' and "L" trials), healthy controls outperformed TL patients (F (2, 38) =5.08, p=.011, r2=.45),

but the difference in score was not significant for controls compared to FL patients. Contrary to

our hypothesis, performance differences between frontal and temporal groups were not

statistically significant on any of the four fluency measures. However, there was a large effect

size for group differences between TL and FL patients on name fluency (d=.75), with FL patients

generating more names than TL patients. Means and standard deviations for patients and control

groups are presented in Table and Figure 4-1.

Repeated-measures ANOVA revealed that both FL (F(3, 18) = 12.37, p<.001) and TL (F

(3, 39)= 30.72, p<.001) patients performed best on semantic fluency, followed by action

fluency, phonemic fluency, and name fluency. FL patients performed significantly better on

semantic fluency as compared to phonemic (p=.005) and name fluency (p=.009). TL patients

performed significantly better on semantic fluency compared to action (p=.002), phonemic

(p=.001), and name fluencies (p=.001). TL patients also performed worse on name fluency

compared to action fluency (p=.01). There was a trend for significant differences between name

and phonemic fluency (p=.06) though this finding did not reach statistical significance (Figure 4-

1).









Aim 2

The second aim of the study was to examine the clinical utility and predictive validity of

experimental fluency measures in identifying seizure location and discriminating between patient

groups.

Before attempting to dissociate patients groups, we sought to determine how well

performance on our panel of fluency measures would dissociate patients from controls. Because

our variables were not significantly collinear (Supermarket: Tolerance=.771, VIF=1.29, Action:

Tolerance=.46, VIF=2.13, Name: Tolerance=.48, VIF=2.10, Phonemic: Tolerance=.51,

VIF=1.95) we used forward-entry logistic regression with all four fluency measures included to

predict patient status. This overall model was significant, and a good fit for the data (X2 (4)

=27.9, p<.001). This model correctly classified seventeen healthy controls as controls, and

classified eighteen patients as patients. The model incorrectly classified three controls as patients

and three patients as controls, resulting in an overall classification accuracy of 85%. Sensitivity

and specificity were both 85%.

We then calculated multiple logistic regressions to determine how well differences in

fluency performance predicted patient group membership. Again, we initially used forward-entry

logistic regression with all four fluency measures included as predictors. Entry of all four

fluency variables into the model revealed a non-significant overall effect (X2 (4) = 7.66, p=. 10)

and a poorly fitted model. This model correctly classified twelve patients with TLE as TLE

patients, but incorrectly classified two TLE patients as having FLE. This model also correctly

classified three FLE patients as FLE patients, though incorrectly classified four FLE patients as

TLE patients. Overall, 86% of TLE patients were accurately classified, whereas 43% of FLE

patients were accurately classified, resulting in a total correct classification accuracy of 71%

(Table 4-3). Because of the poor fit of the model, we subsequently used backwards entry logistic









regression to determine suitability of the variables for the model. Results of this analysis

provided two additional models. The first model included action, name, and semantic fluency,

and showed a trend for significance (X2 (3) = 7.66,p=.053). The final model fit the data well (X2

(2) = 7.66, p=.02), and included only semantic (p=.07) and name fluency (p=.053), removing the

non-significant action (p=.91) and phonemic fluencies (p=.97). This final model correctly

classified twelve TLE patients as having TLE, and incorrectly classified two as having FLE. This

model also correctly classified five patients with FLE as having FLE, and incorrectly classified

two as having TLE, resulting in an overall correct classification rate of 81%. This model

correctly classified 86% of patients with TLE as having TLE, and 71% of patients with FLE as

having FLE. Classification statistics for both models are presented in Tables 4-3 and 4-4.

The predicted probabilities from this final model were subsequently used to predict a

receiver operating characteristic (ROC) curve (Figure 4-2). Using the final predictors in our

model (semantic and name fluency tests only), the overall predicted area under the curve was

.806.

Aim 3

The third aim of the study was to examine the convergent and discriminant validity of the

experimental fluency tests with other tests sensitive to frontal and temporal lobe dysfunction, and

to examine the relationship amongst fluency measures.

We first examined the relationship amongst fluency measures for patient groups only and

then for patients and controls combined. When patients and controls were combined, there were

positive, statistically significant correlations amongst all fluency measures. Semantic fluency

was significantly correlated with name fluency (r=.407,p=.008), action fluency (r=.446,

p=.003), and phonemic fluency (r=.382, p=.01). Name fluency was also significantly positively

correlated with action (r=.662,p<.001) and phonemic fluency (r=.628, p<.001). Performance on









action fluency was also correlated with phonemic fluency performance (r=.637,p<.001). The

correlation matrix for patients and controls is presented in Table 4-6.

When we examined patients alone, there were no statistically significant correlations

between semantic and phonemic fluency (r=.076,p>.05), name and phonemic fluency (r=.144,

p>.05), semantic and action fluency (r=.210, p>.05), or name and action fluency (r=.285, p>.05).

However, there were trends for significance for some correlations. As predicted, semantic and

name fluencies were positively correlated with a moderate effect size (r=.415,p=.06). There was

also a positive correlation of moderate effect between action and phonemic fluencies (r=.390,

p=.08). These results are presented in Table 4-5.

We also examined the relationship between performance on fluency tests and performance

on traditional neuropsychological measures thought to be sensitive to the presence of language-

dominant TL and FL dysfunction. These included the Boston Naming Test (BNT) and Logical

Memory (LM) I and II (WMS-III), and Digit Span (DS) (WAIS-III), Wisconsin Card Sorting

Test (WCST), and Trails B (TMT-B), respectively. Measures of particular interest included the

number of categories correctly completed and number of perseverative responses on the WCST,

and total time to completion and total errors on the TMT-B.

We examined these relationships separately for patients, and then for patients and controls

combined. When patients and controls were combined, there were statistically significant

positive correlations between semantic and name fluency and scores on the BNT, LM-I, and LM-

II, such that better performance on fluency tests was associated with better performance on these

measures ((semantic and BNT: r=.449,p<.01; semantic and LM-I: r=.542,p<.001; semantic and

LM-II: r=.601,p<.001) and (name and BNT: r=.582,p<.001; name and LM-I: r=.624,p<.001;

name and LM-II: r=.638, p<.001)). Semantic fluency performance was negatively correlated









with TMT-B time (r=-.382, p<.05) and TMT-B errors (r=-.455, p<.01). Finally, better name

fluency was significantly associated with better performance on DS (r=-.401,p<=01).

Action and phonemic fluency were also positively correlated with performance on the

BNT, LM-I, and LM-II. Action fluency performance was significantly positively correlated with

DS total score (r=.493,p<.01) and WCST-Categories (r=.429,p=.01). Action fluency was

significantly negatively correlated with TMT-B time (r=-.485,p<.01) and WCST-perseverations

(r=-.381,p<.05), such that persons who performed worse on this test had more perseverations on

the WCST and took longer to complete TMT-B. Phonemic fluency scores exhibited a similar

pattern; significant correlations were found between phonemic fluency and DS total score

(r=.527, p=.001), TMT-B time (r=-.430, p<.01), and WCST-perseverations (r=-.392,p<.05).

When examining patients separately, semantic fluency remained positively correlated with

LM-II (r=.467, p=.05), and there was a moderate positive correlation and trend for significance

with LM-I (r=.416, p=.08). There was also a moderate negative correlation and trend for

significance with TMT-B errors (r=-.424, p=.07). Additionally, action fluency remained

negatively correlated with TMT-B time (r=-.453,p<.05), and positively correlated with BNT

(r=.543,p<.05) and LM-II (r=.476,p<.05). There was a trend relationship between action

fluency and WCST-categories (r=.448, p=.07). Scores on phonemic fluency were positively

correlated with DS total score (r=.494,p<.05), negatively correlated with TMT-B time (r=-.510,

p<.05), and there was a trend for significance for WCST-perseverations (r=-.408, p=. 12) and

BNT (r=.415,p=.08). There were no significant correlations with name fluency. Correlation

matrices are presented in Tables 4-7 and 4-8.

Aim 4

The fourth aim of our study was to determine whether a qualitative analysis of fluency

performance, including the number of clusters, cluster size, and switches, would dissociate









performance of FL and TL patients. Additionally, we sought to ascertain whether TL and FL

patients displayed fewer clusters and switches on fluency tests thought to most sensitive to

temporal and frontal lobe dysfunction, respectively.

We examined the number of clusters, switches, and mean cluster size for each individual

fluency measure separately. On the semantic fluency task, controls, TL patients, and FL patients

generated a similar number of clusters (F (2, 36) = .574, p= .56, 12=.17) and switches (F (2, 36)

= 3.14, p=.06, r12=.38), and did not differ significantly on the size of clusters generated (F (2, 36)

=.57, p=.57, 2=. 17). Controls and patients also performed similarly on measures of phonemic

fluency, with no significant group differences on number of clusters (F (2, 38) = .05, p=.94,

12=.06), switches (F (2, 38) = 2.87, p=.07, 12=.36), or cluster size (F (2, 38) = .40,p=.67,

12=.14). On action fluency clusters, significant group differences were found, (F (2, 36) = 4.1, p

=.025, 12=.43) and planned contrasts revealed that controls generated significantly more clusters

than TL patients (p =.05), but not FL patients. Controls also switched between clusters more

frequently than patients (F (2, 36) = 4.9,p=.01, 12=.45), with planned contrasts revealing

controls switched more frequently than TL patients (p=.01). There were no significant group

differences in action fluency cluster size (F (2, 36) = 2.25, p =.12, r12=.33). On name fluency,

there were significant group differences for total number of clusters (F (2, 36) = 3.1, p=.05,

r12=.38) and switches (F (2, 36) = 7.75,p <.01, r2=.54), and a trend towards significance for

mean cluster size (F (2, 36) = 2.45, p=.10, r2=.34). Planned contrasts revealed that TL patients

generated significantly fewer name clusters than controls (p=.05), and that both TL (p<.01) and

FL (p<.05) patients switched less frequently than controls. While not statistically significant,

there was a large effect (p=. 16, d=.71) showing that TL patients also generated fewer clusters

than FL patients on name fluency. Despite the non-significant finding for group differences in









cluster size, there were also large effect sizes for between-group differences for TL and FL

patients (d=.88), and TL patients and controls (d=.73), with TL patients generating smaller

clusters than both groups. Data for clusters, switches, and mean cluster size across fluency type

are presented in Figures 4-3 through 4-5.

Additional Study Aims

While not the primary aims of our study, we were interested in examining the

relationship amongst performances on naming analogues of our fluency paradigms, and between

naming and fluency performance.

Because scores on naming tests were not normally distributed, we used non-parametric

tests (Kruskal-Wallis and planned Mann-Whitney contrasts) to assess for group differences in

naming performance. Number of actions correctly named on the Action Naming Test differed

significantly by group (H(2) = 19.54, p<.001), with controls correctly naming more actions than

both TL patients (U=21, z=-4.19,p<.001) and FL patients (U=21, z=-2.74,p<.01). Contrary to

our hypotheses, the two patient groups did not differ significantly in their ability to name actions

(U=46, z=-.225, p>.05).

Performance on the Famous Face Naming Test was assessed by computing a "percent

correct" score, which was the total number of items correctly named out of the items correctly

recognized, multiplied by 100. There were significant group differences in the ability to name

famous faces (H(2) =18.95,p<.001). TL patients were able to accurately name only 33% of the

faces they recognized, as compared to 53% for FL patients and 71% for controls. As expected,

both TL and FL patients performed significantly worse than controls on this test (TL: U=22, z=-

4.13,p<.001; FL: U=32.5, z=-2.07,p<.05) and consistent with our hypotheses, patients with

TLE performed significantly worse than patients with FLE (U=22.5, z=-l.97, p<.05).









On a test of common object naming (Boston Naming Test), controls were able to correctly

name more items without cueing than both FL and TL patients (H(2)=17.48,p<.001; TL: U=16,

z=-3.97,p<.001; FL: U=15, z=-2.16,p<.05). On average, controls correctly named 55/60 items,

FL patients named 49/60, and TL patients named 43/60. Naming differences between FLE and

TLE patients were not statistically significant (U=18, z=-1.44, p>.05), but a moderate effect size

(r=.33) suggests FL patients performed better than TL patients.

Spearman rho non-parametric correlations were conducted to examine the relationship

amongst naming measures and between fluency and naming performance. When combining

patients and controls, performance on the three naming tests was highly positively correlated.

Performance on the ANT and BNT was also highly correlated with fluency test performance

regardless of fluency type. Famous faces naming score was significantly correlated with scores

on semantic fluency, name fluency, and action fluency. When we examined patients alone, BNT

score was significantly correlated with ANT (r=.542, p<.01) and FFNT (r=. 508, p=.01) scores,

but there was only a trend relationship between ANT and FFNT scores (r=.314, p=.08). Neither

BNT nor FFNT score was significantly associated with performance on fluency measures.

Performance on the ANT was significantly positively associated with phonemic fluency (r=.505,

p=.01) and name fluency (r=.372,p<.05) scores. Correlations are presented in Tables 4-10 and 4-

11.









Table 4-1. Performance on fluency measures
LTL FL Controls
N 14 7 20
Phonemic (C) 10.1(4.3) 9.60 (2.6) 13.3 (5.4)
C+F+L 27.7 (11.3) 28.7 (4.7) 38.4 (11.1)*
Semantic 18.1 (5.4) 16.5 (4.2) 22.5 (6.5)
Action 11.7(4.5) 12.7 (2.9) 18.3 (5.8)**
Name 6.1 (3.5) 8.8 (2.3) 14.5 (4.7)**
Note: Means are presented, standard deviations in parenthesis.
** p.01 for controls versus LTL and FL
*p<. 05 for controls vs LTL


Table 4-2. Multivariate analysis of fluency performance in patients
Pillai's Trace F p Partial Eta-Squared
Phonemic + Action Model .015 0.14 .87 .015
Semantic + Name Model .280 3.54 .05 .282


Table 4-3. Four fluencies predicting patient group membership
LTL (predicted) FL (predicted) % correct
LTL (actual) 12 2 85.7
FL (actual) 4 3 42.9
71.4

Table 4-4. Semantic and name fluencies predicting patient group membership
LTL (predicted) FL (predicted) % correct
LTL (actual) 12 2 85.7
FL (actual) 2 5 71.4
81.0


Table 4-5. Correlations coefficients for fluency measures (patients only)
(1) (2) (3) (4)


(1) Phonemic Fluency


(2) Semantic Fluency


(3) Name Fluency


1.00


.076
.74

.144
.54


1.00


.415
.06


1.00


(4) Action Fluency .390 .210 .285
.08 .36 .21
Pearson correlation coefficients are presented followed by significance values.


1.00













Table 4-6. Correlations coefficients for fluency measures (patients and controls)
(1) (2) (3) (4)


(1) Phonemic Fluency


(2) Semantic Fluency


(3) Name Fluency


1.00


.388
.01*

.628
.00*


1.00


.407
.01*


1.00


(4) Action Fluency .637 .446 .662
.00* .00* .00*


1.00


Pearson correlation coefficients are presented followed by significance values.
*p<.01


Table 4-7. Correlations amongst neuropsychological measures (patients only)
DS TMT- TMT-B WCST- WCST- BNT LM-I LM-II
B Err. Cat. Pers.
Semantic -.108 -.311 -.424 .034 -.095 .143 .416 .467
.64 .18 .07 .89 .73 .78 .08 .05*

Action .365 -.453 -.285 .448 -.363 .543 .326 .476
.11 .04* .24 .07 .17 .02* .19 .04*

Name .103 .066 -.157 -.171 .077 .124 .233 .240
.66 .78 .52 .51 .78 .62 .35 .33

Phonemic .494 -.510 -.282 .231 -.408 .415 -.290 .095
.02* .02* .24 .37 .12 .08 .24 .70
DS=Digit Span total score; TAT-B Trail Making Test-B time; WCST= Wisconsin Card .', ,,n, Test;
BNT=Boston Naming Test, LM Logical Memory
Pearson correlation coefficients are presented followed by significance values.
*p<.05












Table 4-8. Correlations amongst neuropsychological measures (patients & controls)
DS TMT- TMT-B WCST- WCST- BNT LM-I LM-II
B Err. Cat. Pers.
Semantic .194 -.382 -.455 .191 -.212 .449 .542 .601
.243 .028 .01** .27 .23 .01** .00** .00**

Action .493 -.485 -.260 .429 -.381 .683 .638 .602
.00** .00** .13 .01** .02* .00* .00* .00*

Name .407 -.253 -.224 .161 -.242 .582 .624 .638
.01** .13 .20 .36 .17 .00** .00** .00**

Phonemic .527 -.430 -.285 .222 -.392 .566 .328 .453
.00** .01** .10 .21 .02* .00** .05* .01**
DS=Digit Span total score; TAT-B Trail Making Test-B; WCST Wisconsin Card \N', Test;
BNT=Boston Naming Test, LM Logical Memory
Pearson correlation coefficients are presented followed by significance values.
**p<.01
*p<.05

Table 4-9. Performance on naming measures
LTL FL Controls
BNT 43.4 (6.7) 48.6 (5.0) 54.9 (4.7)a
ANT 47.4(3.7) 48.0(4.1) 53.1 (1.9)a
FFNT 33.3 (21.1) 53.9 (20.4)b 71.9 (16.9)a
Note: Means are presented, standard deviations in parenthesis for
ANT and BNT. FFNT is presented as % correct x 100.
a Controls >TL and FL
bFL>TL

Table 4-10. Correlations between naming and fluency measures (patients and controls)
Phonemic Semantic Name Action
FFNT ANT BNT Fluency Fluency Fluency Fluency
FFNT 1.00 .651 .664 .234 .368 .433 .350
.00* .00* .07 .01* .00* .01*
ANT .651 1.00 .853 .413 .513 .565 .489
.00* .00* .00* .00* .00* .00*
BNT .664 .853 1.00 .403 .456 .596 .628
.00* .00* .01* .00* .00* .00*
Pearson correlation coefficients are presented followed by significance values.
*p<.01











Table 4-11. Correlations between naming and fluency measures (patients only)

Phonemic Semantic Name Action
FFNT ANT BNT Fluency Fluency Fluency Fluency


FFNT 1.00 .314 .508 .136 .177 .003
.08 .01** .27 .22 .49

ANT .314 1.00 .542 .505 .283 .372
.08 .01** .01** .10 .04*

BNT .508 .542 1.00 .265 .056 .013
.01** .01** .14 .41 .47

Pearson correlation coefficients are presented followed by significance values.
**p<.01
*p<.05


25




20

-





E
S150


I-
1--



5-




0
-


.116
.30

.191
.20

.363
.06


Semantic Phonemic Action Name
Control > LTL and FL at p<.05


Figure 4-1. Overall fluency performance across groups


-L TL
-W- FL
C ,- ,Ir



























0.6-
>


A 0.4-



0.2-



0.0- I I
0.0 0.2 0.4 0.6 0.8 1 .C

1 Specificity

Figure 4-2. Receiver operating characteristic (ROC) curve for semantic and name fluencies
predicting patient group





















I I
II


15















05







0
Phonemic Semantic
significantt at p<05

Figure 4-3. Mean number of clusters by group





20


Phonemic Semantic
significant at p< 05

Figure 4-4. Mean number of switches by group


I I


Action Name


Name


Action


HIITL

0 dZ-I-






















3




25-




2



NFL
15



1-




05




01
Phonemic Semantic Action Name



Figure 4-5. Mean cluster size by group









CHAPTER 5
DISCUSSION

Summary of Findings

The current study was undertaken to help understand differences in cognitive impairment

in patients with left temporal lobe epilepsy versus epilepsy localized to the frontal lobes. More

specifically, we sought to elucidate a distinct pattern of fluency test performance that would

discriminate between these two patients groups. We were interested in examining differences in

both traditional semantic and phonemic fluency performance, as well as performance on two

experimental fluency tests, action and name fluency. Scientific literature positing different neural

networks for retrieval of action words and proper names suggested that incorporation of this

material-specific content into traditional fluency test paradigms would improve measurement of

the unique cognitive deficits associated with localized neural dysfunction. Furthermore, we

hoped that an examination of fluency strategy, including generation of clusters, cluster size, and

switches between clusters, would provide useful diagnostic information about our patients. In

other words, by manipulating the fluency retrieval demands involved and examining the

cognitive strategies employed, we hoped to more accurately discriminate between patient groups,

in addition to advancing our understanding about the neural specificity of the brain regions

involved.

The first primary aim of this study was to characterize performance of patients with frontal

or left temporal lobe epilepsy, and matched healthy controls on a panel of verbal fluency tests

that included clinical measures of semantic and phonemic fluency, and experimental measures of

action and proper name fluency. We hypothesized that patients with epilepsy would perform

worse on all of these measures than would our healthy controls. We also believed that patients

with temporal lobe epilepsy would evidence impaired semantic fluency (but not phonemic









fluency), while patients with frontal lobe epilepsy would demonstrate impairments on phonemic

fluency (but not semantic fluency). With regard to our experimental fluency measures, we

believed that overall fluency score on action and proper noun fluency would doubly dissociate

patients with frontal and temporal lobe epilepsy, with frontal lobe patients performing worse on

action fluency and temporal lobe patients exhibiting comparative deficits on tests of proper name

fluency. These hypotheses were only partially confirmed. As expected, controls generated more

words across fluency tests than both patient groups. Interestingly, however, the findings were

statistically significant only for the experimental action and name fluency tests, with controls

outperforming patients with both types of epilepsy. While there were trends for significance and

moderate effect sizes for patient and control group differences on semantic and phonemic

fluency, these effects did not reach a level of statistical significance. Contrary to our predictions,

there were no statistically significant differences between the two epilepsy groups on any of the

fluency measures. However, there was a moderate effect size for name fluency, with TL patients

generating fewer proper names than FL patients, suggesting meaningful group differences that

could not be adequately detected because of our sample size. In both patient groups, participants

generated the most words for supermarket fluency, followed by action, phonemic, and name

fluency, a finding consistent with past literature (Piatt et al., 1999).

Our second aim was to determine the predictive validity of traditional and experimental

fluency measures in predicting patient group membership. We predicted that both traditional and

experimental fluency measures would adequately discriminate between patients with frontal and

temporal lobe epilepsy, and that the multivariate combination of our four fluency measures

would have superior predictive ability above and beyond either traditional or experimental

fluency measures alone. These hypotheses were not confirmed in our study population. We









found that our fluency panel did an inadequate job of accurately predicting patient group

membership. Our four-fluency model better predicted temporal lobe than frontal lobe

involvement, but was still a poor fit for our data. Our final model, which fit the data well,

included only semantic and name fluency and accurately predicted 86% and 71% of our TL and

FL patients, respectively. This suggests that the tests removed from the model (action and

phonemic fluency) did not offer additional predictive value beyond the variables in our final two-

predictor model.

The third aim was to examine the convergent and discriminant validity of our experimental

fluency measures using traditional neuropsychological measures of frontal and temporal lobe

functioning as criterion variables. We hypothesized that performance on tests of common and

proper noun fluency would be more related to measures also sensitive to the integrity of the

temporal lobe (i.e., related to semantic stores), while action and phonemic fluency scores would

exhibit small to moderate relationships with traditional measures of executive function. When

examining performance for controls and patients combined, we found strong positive

correlations amongst all four fluency measures, suggesting the presence of a common source of

variance in fluency performance regardless of the retrieval demands. When only patients were

included in the analysis, phonemic and action fluency were moderately positively correlated, as

were name and semantic fluency. Name and semantic fluency were uncorrelated with both action

and phonemic fluency, suggesting that in patients with localized neuronal dysfunction, fluency

performance was differentially impaired based on the category-specificity of the material to be

retrieved.

When we examined relationships between our fluency measures and other measures of

neuropsychological functioning for our patients and controls, there were significant positive









correlations between all four measures and scores on tests of language and verbal memory (BNT,

LM-I and LM-II). This was consistent with our predictions for name and semantic fluency, but

not action and phonemic fluency. As predicted, action and phonemic fluency were correlated

with measures of executive function (Digit Span number correct, TMT-B time, and WCST-

perseverations). Contrary to our prediction, semantic fluency was correlated with performance

on TMT-B (time and errors) and name fluency was related to performance on Digit Span.

The fourth aim of the present study was to determine whether a qualitative analysis of

fluency performance, including number of clusters, switches, and cluster size, would dissociate

performance of patients with FLE, TLE, and healthy controls. Based on existing literature, we

hypothesized that patients with TLE would exhibit reduced cluster size, particularly on tests that

carry a heavier semantic burden (name fluency, semantic fluency) and that patients with FLE

would evidence a reduced number of switches, primarily on tests of phonemic and action

fluency. These hypotheses were only partially confirmed. There were no significant group

differences for clusters, switches, or cluster size for phonemic or semantic fluencies. On action

fluency clusters, controls generated more clusters than did either patient group, though this was

statistically significant for only for the control versus TL comparison. Controls also switched

more frequently than TL patients. There were no significant group differences in action fluency

cluster size. On name fluency, TL patients generated significantly fewer name clusters than

controls, and both TL and FL patients switched less frequently than controls. There was a large

effect indicating TL patients also generated fewer clusters than FL patients. Large effect sizes

also revealed that TL patients generated smaller clusters than FL and controls on name fluency,

which is consistent with our a priori hypothesis.









The body of naming literature posits distinct neural substrates for action naming (i.e. the

frontal cortices) and person naming (i.e. the anterior temporal cortices of the left hemisphere),

and while there is preliminary support for a comparable fluency substrate, these paradigm have

not been used extensively in the literature. Because of this, we were interested in examining the

relationship between our fluency measures and related naming measures, including the Boston

Naming Test, Action Naming Test, and Famous Faces Naming Test. We hypothesized TL

patients would exhibit more prominent naming deficits than FL patients and controls on

measures of common object naming and famous faces naming, but that patients with FLE would

exhibit deficits on our measure of action naming. Our naming hypotheses were also only

partially confirmed. As expected, controls were able to correctly name more items on all three

naming measures than both FL and TL patients. On the BNT, no statistical differences were

found between patient groups, but a moderate effect size suggested that TL patients were more

impaired at naming common objects than FL patients. TL patients were also worse at naming

familiar famous faces as compared to patients with FL epilepsy; in fact, they named 20% less

than patients in the FL group. Contrary to our predicted results, FL and TL patients did not differ

in their ability to name actions on the ANT.

When examining the relationship amongst naming measures, performance was highly

correlated for patient and controls. Performance on the ANT, BNT, and FFNT was also highly

correlated with fluency test performance regardless of fluency type, except for FFNT and

phonemic fluency. When we examined patients alone, BNT score was significantly correlated

with ANT and FFNT, but no significant relationship was established between ANT and FFNT.

Neither BNT nor FFNT score was significantly associated with performance on fluency

measures. Performance on the ANT was associated with phonemic and name fluencies.









In sum, while many of our hypotheses were confirmed in the present study, a significant

portion were also disconfirmed, particularly as it relates to performance on traditional measures

of fluency, action fluency, and our qualitative analysis of fluency performance. Interpretations of

our study findings are presented below.

Interpretation of Findings

Semantic and Phonemic Fluency

We found that our patient groups did not show differential phonemic and semantic fluency

performance and that neither measure was a good predictor of patient group membership. These

findings were inconsistent with our hypotheses, but not entirely surprising, and part of the reason

we undertook the present study comparing our experimental tests to these traditional measures.

While there are many studies that show successful semantic fluency performance relies more on

the integrity of the left temporal lobe and phonemic fluency is sensitive to the presence of frontal

lobe pathology, there are numerous studies that fail to show this effect. Thus, the current results

are in good company.

Studies have shown equivalent test performance on semantic fluency in patients with

anterior and posterior lesions. In an epilepsy population, Drane et al. (2006) found that patients

with frontal lobe seizure foci were more impaired than a group with temporal lobe epilepsy on

measures of semantic fluency, contrary to their hypotheses that predicted more impairment in the

temporal lobe group. Another study comparing patients with focal anterior and posterior lesions

found that both types of lesions produced impairments on semantic, or category, fluency (Stuss

et al., 1998). Additional studies employing a variety of populations have found a similar pattern

of equivalently impaired semantic fluency in frontal and temporal lobe patients (Baldo &

Shimamura, 1998; Costello & Warrington, 1989; Owen et al., 1990; Randolph et al., 1993).

These authors have argued that rapid word generation, regardless of retrieval demand, can be









impaired by frontal lobe lesions. These studies purport that patients with executive dysfunction

are unable to perform effective, strategic, and efficient searches for words, irrespective of

whether the search is semantically or phonemically driven (Baldo et al., 2006; Baldo &

Shimamura, 1998; Troyer et al., 1998).

A similar pattern has also been established for phonemic fluency tests (Stuss et al., 1998;

Miller, 1984; Pendelton et al., 1982; Perret, 1974). Emory and Alvarez (2006) found that the

bulk of studies in their meta-analysis of frontal-lobe lesion patients reported significantly poorer

phonemic fluency scores compared to controls, however, a significant percentage found similar

impairment on phonemic fluency in patients with non-frontal lobe lesions. Henry and Crawford

(2006) found that phonemic fluency deficits were largest in patients with left frontal lesions, but

that patients with non-frontal left hemisphere lesions were often similarly impaired, suggesting

phonemic fluency performance may be determined both by an executive factor and a verbal

component. In fact, this theory was put forth decades ago by Ramier and Hecaen (1970), who

hypothesized that successful performance on phonemic fluency is determined by an "executive"

factor located within the frontal lobes and a "verbal" factor mediated more generally by the

language-dominant hemisphere.

Equivalently impaired performance on semantic and phonemic fluency in patients with

frontal and temporal lesions suggests that these tests are sensitive to the presence of

frontotemporal damage, but not specific to more localized impairment within this region.

Adequate performance on semantic and phonemic fluency tasks is likely multi-factorial, and may

depend on verbal contributions from the language dominant hemisphere, efficient search and

retrieval strategies dependent on frontal lobe functioning, and a general cognitive factor, or "g".

This was reflected in our study; semantic and phonemic fluency performance was significantly









correlated with measures of verbal/semantic ability but was also related to measures thought to

tap an executive component. Further, WASI Full-Scale IQ was a strong predictor of fluency

score (R2=.26, and .30 for semantic and phonemic fluency) suggesting that fluency performance,

regardless of the retrieval demands, could also be nonspecifically depressed in both groups due

to an overall cognitive impairment associated with chronic, uncontrolled seizures (Jokeit &

Ebner, 2002).

That being said, intact verbal and semantic memory abilities and efficient search/retrieval

strategies likely contribute in different ways to performance on these measures, with the former

being more important to semantic fluency performance and the latter to phonemic fluency

(Butters et al., 1987; Gleissner & Elger, 2000; Janowsky et al., 1989; Jurado, et al., 2000; Martin,

Loring, Meador, & Lee, 1990; Monsch et al., 1992; N'Kaoua, 2001; Rosser & Hodges, 1994;

Stuss et al., 2000; Troster et al., 1995; Troyer et al., 1998). This implies that the original

hypotheses pertaining to our patient groups may be valid, but due to the nature of our study

population, were unable to be borne our within the constraints of our current study. Power

analyses based on data from Troyer and colleagues (1998) suggested that between-group

differences could be detected on semantic and phonemic fluency with a sample of seven-to-ten

patients with left-lateralized frontal and temporal lesions. While studies do not provide

conclusive evidence that phonemic and semantic fluency performance can be doubly-dissociated,

the discrepancy between our findings and those projected are most likely due to differences in

our sample populations. Many studies that have found group differences in TL versus FL

populations, including Troyer et al. (1998), included patients with circumscribed lesions from

stroke and tumors, in addition to etiologies such as traumatic brain injury and seizure surgery.

Similar to our study, her study included a mixed population of relatively acute (i.e., post-stroke)









and chronic patients (i.e., post-surgical intractable epilepsy). Contrary to ours, however, her

patients all had stable (> months post-injury) and focal (dorsolateral prefrontal cortex; superior

medial frontal; inferior medial frontal) lesions. Our study population was different from this in

many regards. A pre-surgical epilepsy population likely suffers from both diffuse (e.g., a history

of intractable seizures with propagation to surrounding brain regions) and focal (e.g., localized

onset) impairments, and can experience overall cognitive depression due to epilepsy medications

or the duration of their disease (Jokeit & Ebner, 2002; Loring, Marino, & Meador, 2007;

Nichols, Meador & Loring, 1993). Additionally, most of the pre-surgical patients in our study

were actively experiencing seizures, which contributed to the chronicity and pervasiveness of

their cognitive dysfunction, making their pattern of neuronal damage dramatically different that

patients who experience an acute injury such as a stroke.

Furthermore, due to recruitment constraints, our study included both pre- and post-surgical

epilepsy patients and patients with both right and left frontal lesions, both of which are factors

that could explain our lack of predicted group differences. By including both pre- and post-

surgical patients, some of whom had circumscribed, defined, and stable lesions and some of

whom did not, we introduced more variance into our study sample. The inability to include only

left frontal patients in our study was possibly the single biggest explanatory factor for our non-

significant findings. The left frontal lobe, particularly the dorsolateral prefrontal region

(DLPFC), has been shown to be the most critical region to phonemic fluency performance

(Milner, 1964; Pendelton et al., 1982; Perret, 1974; Stuss & Levine, 1998), while the right

DLPFC appears to be a less important region. Patients with right DLPFC damage (Miceli et al.,

1981; Miller, 1984; Ramier & Hacean, 1970; Troyer et al., 1998) display impaired phonemic

fluency, but to a lesser degree than their left hemisphere counterparts. There is some evidence









that the right DLPFC contributes to "on-task" behavior important to fluency performance, such

as monitoring, retrieval success, and inhibition of extraneous information (Cabeza and Nyberg,

2001). However, the left dorsolateral prefrontal regions are also thought to play a key role in

these same processes in addition, perhaps, to its preferred access to the lexicon. Imaging data

supports primary involvement of the left DLPFC, and secondary involvement of right frontal

structures, in phonemic fluency performance (Frith et al., 1995; Parks et al., 1988). Damage to

the superior medial frontal regions in both the right and left hemisphere can impair fluency

performance (Stuss & Levine, 2002; Troyer et al., 1998) but are again thought to play a

secondary role to the DLPFC. Unfortunately, the nature of our patient group did not allow for

more precise localization of pathology within particular sectors of the frontal lobe. However,

because of our mixed sample of left and right frontal lobe epilepsy patients, the demands of

phonemic fluency tests (i.e., search and retrieval, organization of unstructured orthographic

information, flexibility) may not have been as taxing as they might have been in a more pure

sample of left-lateralized patients, thereby reducing the extent to which impairments were found.

Action Fluency

Action fluency is a relatively new test construct that grew out of the naming literature in

agrammatic aphasics showing impaired retrieval of words denoting actions in the presence of

spared object naming (Miceli, 1984). The lesion literature also supports deficits in action naming

associated with damage to the frontal cortices. For instance, Damasio & Tranel (1993)

demonstrated a double dissociation in the performance of patients with anterior temporal cortex

damage (who had difficulty naming pictures of objects) compared to another patient with left

premotor damage (who was unable to name actions depicted in line drawings). Deficits in action

naming have been inconsistently demonstrated in patients with fronto-temporal dementia and









various other lesions (Damasio & Tranel, 1993; Monsch et al., 1992; Ostberg et al., 2005; Silveri

et al., 2003).

Action fluency, on the other hand, has been relatively unstudied until recently. Preliminary

studies with HIV and Parkinson's dementia patients have demonstrated impairments in action

fluency compared to semantic and phonemic fluency. These studies have also supported the idea

that action fluency is a construct of executive functioning (Piatt et al., 1999a and 199b; Woods et

al., 2005a and 2005b). These findings were only partially replicated in our patients with frontal

epilepsy. We found that frontal and temporal lobe epilepsy patients performed similarly on both

action word naming and action fluency. As a result, action fluency score was a poor predictor of

patient group membership. We did find support for the notion that, at best, action fluency may

indeed be a construct sensitive to "executive functioning", or that it at least had some

relationship with other measures purported to be sensitive to executive function. Action fluency,

more so than phonemic fluency, was strongly related to performance on the WCST-Categories,

WCST-Perseveration, Digit Span, and Trails B time, providing evidence of convergent validity

amongst measures of executive functioning. Also noteworthy, however, is the fact that action

fluency scores were moderately correlated with measures of verbal/semantic ability, including

the BNT and WMS-LM II. This has not been found previously in the literature, but has also not

been explored fully as Woods et al. (2005) did not include measures of naming or verbal

retrieval in their analysis of correlates of action fluency. Our findings, in conjunction with

previous findings of Woods and Piatt, indicate that adequate performance on this test is multi-

determined, related both to executive functioning abilities and to verbal abilities. Neuroimaging

studies also support the notion that both anterior (frontal operculum, left premotor, left

prefrontal, left insula) and posterior (left mesial occipital cortex, left supramarginal and posterior









temporal regions) cortices may play a role in the retrieval of words denoting actions (Damasio et

al., 2001; Tranel et al., 2001).

Impaired action fluency performance in our temporal lobe group may be in part, attributed

to this factor. Another possible explanatory factor is that temporal lobe epilepsy patients

commonly exhibit difficulty with components of executive functioning (response disinhibition,

impulsivity, set loss, and difficulties with mental flexibility and abstract thinking) secondary to

propagation of seizure related 'neural noise' from temporal to frontal regions via the medial and

lateral limbic circuits (Hermann & Seidenberg, 1995). Moreover, a growing body of literature

suggests that structural and functional abnormalities in TLE patients exist not only within TL

structures, but also in regions outside of the temporal lobes. For instance, significant white

matter changes have been demonstrated in extratemporal cortex, including the frontal lobes

(Hermann et al., 2003; Oyegbile et al., 2006). This pattern of impaired executive functions has

been well documented on the WCST, TMT, and Stroop paradigms, amongst others (Hermann,

Wyler, and Richey, 1988; Martin et al., 2000; McDonald et al., 2005; Trennery & Jack, 1994;

Corcoran & Upton, 1993). In fact, many of these studies have found patients with language

dominant temporal lobe epilepsy to be equally or even more impaired than those with frontally

mediated seizures. This explanation has also been used in part to explain similarly impaired

performance on phonemic fluency tests, and may also extend to our tests of action fluency.

Again, the possibility remains that our temporal lobe patients with longstanding seizure disorders

may also have exhibited depressed cognitive profiles on multiple cognitive domains due to the

cumulative effect of uncontrolled seizures (Jokeit & Ebner, 2002).

Additional differences in our study population and those of Woods and Piatt may help to

explain our discrepant findings. The action fluency construct has been used only in studies of









patients with frontal and subcortical disease, namely HIV and Parkinson's Disease. Both

populations can exhibit significant executive dysfunction, but also deficits in motor and

cognitive processing speed, which are not necessarily hallmarks of focal seizure disorders. The

impact of cognitive slowing on fluency performance was not accounted for in studies by Piatt or

Woods, despite significant relationships between action fluency scores and scores on measures

of cognitive and motor processing (Woods et al., 2005). It may be the case that the combination

of executive dysfunction and cognitive slowing in these populations differentially affected action

fluency performance compared to semantic fluency, for instance, which could help explain the

intact performance of our frontal lobe epilepsy patients. This could be the case for action fluency

in particular given the hypothesized "executive" burden of the test. Finally, as with semantic and

phonemic fluency performance, our mixed sample of left and right frontal patients probably

reduced our ability to find a significant effect that may have otherwise been present in a solely

left-frontal sample.

With regard to the two predominant theories that exist to explain the discrepancy between

retrieval of action versus object words, the first states that knowledge about objects and actions is

stored in association cortices adjacent to the primary cortical regions that process these classes of

stimuli (Damasio & Tranel, 1993; Perani et al., 2009). As such, object knowledge is stored in

cortical regions adjacent to the occipito-temporal visual stream, while action knowledge is stored

adjacent to structures in the frontal lobe including the prefrontal cortex, premotor cortex, and

supplementary motor area and focal lesions to these areas can disrupt successful retrieval of

action words. The second theory holds that the deficit is largely executive in nature, and relates

to the difficulty of "mentally coordinating and manipulating the large amount of information

related to action-words" (de N6brega, Nieto, Barroso, & Mont6n, 2007; Grossman, 1998; Silveri









et al., 2003; White-Devine et al., 1996). This latter theory also posits that as with all fluency

paradigms, verbal ability may mediate overall fluency score regardless of executive capacity.

In reality, these theories are not mutually exclusive and the scientific literature provides

support for both. It is likely the case that while acute lesions to premotor or association motor

cortices disrupt successful retrieval of action words, insult to other neuronal regions or pathways

is also sufficient to impair this ability. The current study provides greater support for the second

theory, though the former cannot be tested fully because our group was not comprised of patients

with focal lesions in the aforementioned regions.

Name Fluency

Generally speaking, our findings with regard to performance on tests of proper name

generation were consistent with our hypotheses. Name fluency proved to be the most demanding

fluency measure for all patients and controls, but TL patients were differentially impaired on this

test, suggesting a true deficit in this ability rather than a main effect of task difficulty. Patients

with left-temporal lobe epilepsy, both pre- and post-surgical, showed the weakest performance

on this measure, generating on average only six accurate responses in the span of a minute,

which was statistically different than controls, and different from FL patients based on measures

of effect size. Indeed, name fluency was the strongest predictor of group membership in our

regression analysis. In fact, classification accuracy statistics for name fluency alone were rather

comparable to the model that additionally included semantic fluency, with the two-predictor

model correctly classifying only two additional TL patients.

As we expected, in our patient groups, name fluency was related to the traditional measure

of semantic fluency, but not to measures of action or phonemic fluency, providing evidence for

the convergent and discriminant validity of the measure as closely allied with semantic retrieval.

When we examined patients and controls combined, our correlations show that this measure is









strongly related to other tests assessing verbal and/or semantic abilities such as the BNT, LMI

and LMII, reflecting good external validity for a measure thought to be sensitive to temporal lobe

functioning. However, for patients alone, no significant relationships with measures of language

or memory, both largely mediated by language-dominant temporal structures, were seen. While

surprising initially, this finding is consistent with studies showing that deficits on famous face

naming tests may be dissociated from deficits on measures of common object naming (Drane et

al., 2008). In other words, this test construct may be mediated less by general verbal abilities

than most fluency tests, and may rely more upon a unique "semantic" factor not tapped by other

assessment instruments. This feature may make famous face naming particularly sensitive and

specific to anterior temporal lobe functioning. A roughly equivalent pattern of results was

exhibited when we embedded this test construct in a confrontation naming paradigm, lending

further support to our hypothesis that generation of proper names is dependent on the integrity of

temporal lobe structures. For patients and controls, significant relationships were found amongst

the FFNT and all other fluency and naming tests, but when we examined patients alone, the

relationship with the BNT emerged as the only significant relationship, followed by the weak

correlation with the ANT. On this famous faces naming measure, temporal lobe epilepsy patients

were able to correctly generate names for only 1/3rd of the familiar faces, compared to roughly 12

and 3% for the frontal lobe patients and controls, respectively.

Our findings provide support for the notion that the anterior portion of the left temporal

lobe plays a critical role in the ability to generate names or apply labels to people or objects, and

is particularly important in the case of proper names. Existing lesion studies (Fukatsu et al.,

1999; Glosser, Salvucci, & Chiaravalloti, 2003; Martins & Farrajota, 2007) and functional

imaging data (Gorno-Tempini et al., 1998; Grabowski et al., 2001; Tranel, 2006; Tranel,









Grabowski, Lyon, & Damasio, 2005; Tsukiura et al., 2002) provide support for this hypothesis.

This area has been deemed a "convergence zone" by Damasio, Tranel, and colleagues (Damasio,

et al., 1996; Damasio et al., 2004; Tranel, Damasio, & Damasio, 1997;Tranel et al., 2003).

These authors propose that the anterior temporal lobe serves as a region that helps bind multi-

modal sensory and motor input from our surroundings, leading to the development of amodal

concepts. Since names are arbitrary labels that denote members of conceptual categories,

damage to this area can produce category-specific deficits in word retrieval.

Specific deficits have been shown for animals and "unique entities" such as people and

landmarks (Damasio, 1996; Fukatsu et al., 2000; Glosser, Salvucci, & Chiaravalloti, 2003;

Gomo-Tempini et al., 1998; Grabowski et al., 2001; Milders, 2000; Tranel, 2006). The apparent

difficulty with retrieval of the latter has to do with the "semantic uniqueness" of the object or

person (Semenza & Zettin, 1989; Glosser, Salvucci, & Chiaravalloti, 2003; Grabowski et al.,

2001; Tranel, 2006). Whereas common names refer to concepts, or a set of attributes that are

shared by multiple entities within the same concept, proper names do not inherently contain

attributes in and of themselves and are merely expressions by which we refer to an individual

person or item. Because of this, it is thought that widespread neural networks support the

representation of common nouns, while proper nouns are thought to hold rather fragile

"associations" with their unique reference (Gorno-Tempini & Price, 2001; Martins & Farrajota,

2007). This distinction is particularly salient when contrasting between generation of common

versus proper names on fluency and naming tests, and may explain why the latter are emerging

as particularly sensitive to anterior temporal lobe damage. For instance, the semantic

representation for the word "dog" (an appropriate response for the semantic fluency category

"Animals") is likely much more substantial, perhaps encompassing the words "beagle", "pug",









"dalmation", "poodle", and the like, whereas "Bob Dole", "Marilyn Manson", "Hulk Hogan",

and "Mother Theresa" all refer to singular, unique entities, not linked in cohesive semantic

networks.

Our findings suggest that a name fluency paradigm may offer particular value in detecting

the type of impairment that is common in both pre- and post-surgical language-dominant

temporal lobe epilepsy patients. Name fluency score appears to be less affected by frontal lobe

fluency processes such as efficient monitoring/searching/flexibility than are other measures of

fluency, including semantic fluency, and are more contingent upon adequate functioning of the

temporal lobe semantic networks. It may also be the case that a name fluency paradigm could

reveal impairments that are not evident on other types of neuropsychological tests. Drane and

colleagues (2008) have described patients with subjective complaints of post-surgical naming

deficits who perform at expectation on measures of common object naming but show impairment

on their famous-faces naming test. This suggests that traditional clinical measures, which focus

only on object naming, may not adequately tap the type of abilities commonly disrupted by TLE

or anterior-temporal lobectomy.

Qualitative Analysis of Fluency Performance

Unfortunately, we did not find that an analysis of qualitative fluency performance provided

much additional useful information about the patient groups in our study. In general, the pattern

of qualitative fluency performance echoed the quantitative analysis of performance; no

significant group differences were found on measures of semantic and phonemic fluency. TL

patients generated fewer clusters and switches on action fluency than controls, and fewer and

smaller clusters than controls and FL patients on name fluency. Both patient groups switched

less frequently than controls on name fluency; all results that mirror the quantitative impairments

just discussed.









This raises the following question: are number of clusters, cluster size, and switches

actually proxies for overall fluency performance or are they independent measures of the

functional integrity of the temporal and frontal structures? There is evidence to support the

former (proxy) view, as qualitative measures of fluency have been shown to be highly tied to

overall fluency score in an Alzheimer's and Parkinson's sample (Troyer et al., 1997). This

pattern was replicated in our data; measures of qualitative performance, across group and fluency

task, were correlated with overall number of words generated. Most studies of clustering and

switching as measures of temporal and frontal lobe functioning have found a similar pattern

(Troyer et al., 1998a, 1998b; N'Kaoua et al., 2001). When studies have found overall group

differences in fluency score, they also have reported differences in clustering and switching

across tasks. When group differences were not found for overall total score, differences did not

tend to emerge on clustering and switching analyses. Troyer addresses this issue directly

(1998a), and argues against this point, but does recognize that switching score and words

generated may be correlated variables because the "number of words generated were always

associated with group differences in switching." In earlier work, she provides direct evidence

that decreased clustering does lead to an overall decrease in total words produced (Troyer et al.,

2007). It is certainly plausible that two patients with the same overall fluency score often, for

instance, could produce very different patterns of performance (i.e., orange-grape-apple-cheese-

yogurt-butter-paper towel-napkin-toilet paper versus orange-hot dog-paper towel-pencils-wine-

beer-champagne-cat food-cheese-carrots), thereby achieving differing scores on clustering and

switching. However, the data from our study and others finds this scenario less likely when

differences in overall score are not apparent.









An alternative explanation is that clusters and switches are not necessarily proxies of

overall fluency score, but that clusters/switches and overall score are both proxies of temporal

and frontal lobe functioning. Under this premise, the lack of significant findings across most of

qualitative (and quantitative) measures has to do with the underlying level of localized neuronal

dysfunction in our patient groups. Again, the heterogeneous nature of our sample, our frontal

lobe group in particular, may have lessened our ability to replicate the effects found in other

studies of patients with more circumscribed lesions of the left or bilateral frontal lobes. Troyer

and colleagues (1998a) were able to compare performance across subgroups of frontal patients

(i.e., left dorsolateral prefrontal cortex (LDLPFC), right dorsolateral prefrontal cortex

(RDLPFC), superior medial frontal (SMF) cortex, inferior medial frontal (IMF) cortex) and

found between-group differences that support this assertion; specifically, switching was impaired

in the LDLPF and combined SMF groups, but not the RDLPF and IMF groups. This finding is

consistent with other neuroimaging and cognitive studies showing impaired initiation of

behavior, poor cognitive flexibility, and perseverations are most strongly related to the left

dorsolateral frontal, inferior frontal, and anterior cingulate regions (Hirschorn & Thompson-

Schill, 2006; Troster et al., 1998). Most likely, our sample included patients whose damage

transcended these functional boundaries, muddying any profile that may have resembled those

previously reported with regard to switches in particular.

Despite our non-significant findings between patient groups with regard to switches, and

clusters on action, phonemic and semantic fluency, the reduced cluster number and size on name

fluency in our temporal lobe group is an interesting finding, and consistent with our a priori

hypothesis. Unfortunately, our small sample size kept these numbers from reaching statistical

significance, but our large effect sizes suggest they are indeed meaningful findings. Our results









show that not only did patients with left temporal lobe epilepsy have difficulty generating proper

names, they also had more difficulty than frontal lobe patients or controls in linking response

items to each other in a semantically meaningful way. When they were able to successively

generate semantically-related names, they tended to be able to generate fewer of these names

before exhausting the semantic network in which they reside. These findings on name fluency

are particularly interesting, as this was the measure we hypothesized to be most sensitive to

impairment of semantic networks subserved within anterior temporal structures. Consistent with

our hypotheses pertaining to overall name fluency performance, this reduced ability to generate

semantically related proper names in particular may reflect the disruption or degradation of

semantic memory stores within the temporal lobes. The finding of reduced clusters and cluster

size on tests sensitive to temporal-lobe functioning has been reported frequently; Troyer has

hypothesized that the best indices for discriminating patient groups were "phonemic-fluency

switching... and semantic-fluency clustering" (Troyer et al., 1998a, 2000; Reverberi, Laiacona,

and Capitani, 2006). This observation indicates that a combination of overall fluency score and

an analysis of clustering and switching by fluency type may provide useful information about the

underlying cognitive impairments in various patient groups.

Limitations of the Present Study

There were a number of factors that limited the present study and may have affected our

ability to find the results we predicted. We have previously discussed most of these, but they will

be reviewed herein. First and foremost, our small heterogeneous sample likely negatively

impacted our study. Unfortunately, even with an extended recruitment time of eighteen months,

we were unable to collect data on enough pre-surgical patients with left lateralized temporal and

frontal lobe epilepsy. This may have been due to a number of factors, though the primary reason

was decreased patient flow through Shands inpatient epilepsy monitoring unit. To supplement









patient flow, we concurrently recruited post-surgical epilepsy patients who had undergone

surgery between 2000 and 2007. By employing this recruitment strategy, we were able to meet

our expected sample size for our left-temporal group. However, we were still not able to recruit

enough participants with left frontal epilepsy and in fact, no addition left frontal patients were

able to be recruited through post-surgical mailings. The alternative to this solution was to include

patients with frontally-localized epilepsy, regardless of laterality. Through this means, we were

able to meet our projected sample size in both patient groups.

However, this likely limited our ability to test out our hypotheses as they pertained to left

frontal-lobe functioning in particular. Many of our predictions with regard to tests of frontal-lobe

functioning (i.e., phonemic and action fluency) were based on the notion that the left frontal

region, the DLPFC specifically, plays a key role in successful fluency performance. This could

not adequately be explored in our current sample because of the mixed nature of our group.

While other areas of the frontal lobes, including the bilateral portions of the superior medial

frontal lobe may also be involved in fluency performance, the lack of specificity in our sample

prohibited us from testing the contribution of various regions to task behavior. Nonetheless, we

did still find interesting differences related to some of our study hypotheses, largely related to

tests sensitive to temporal-lobe functioning, which were less impacted by the heterogeneous

nature of this sample. Many of these findings still only reached clinical, not statistical

significance, implying our study may still have been slightly underpowered.

That being said, the pre-surgical patients we screened and recruited into our study were

consecutive admissions to a clinical epilepsy center and are representative of the actual type of

patients neuropsychologists are asked to assess for pre-surgical evaluations. Patients with

epilepsy often do not have clearly localized seizures. They may have multiple seizure foci,









seizures that propagate from one brain region to another, a mixed pattern of focal and

generalized events, or have a mixed profile of electrographic and non-epileptic seizures, all of

which can complicate the localization/lateralization process. These patients also frequently

present with psychiatric illness, past surgeries, significant head injury, and comorbid diagnoses

of learning disability or mental retardation, and take multiple medications, all obscuring the

diagnostic picture even farther. So in reality, the sample with which we tested our measures and

hypotheses was not ideal, but in many ways, most representative of the type of patient on which

these measures would be used clinically. While this limits our ability to make conclusions about

the sensitivity and validity of our measures, it provides useful information about how a typical

patient with frontally-mediated seizures might perform.

One of the initial points of the current study was to assess the utility of our standard

fluency measures and develop new measures more sensitive to the presence of frontal or

temporal lobe dysfunction. This is particularly important as it relates to our ability to provide

useful information to epileptologists and epilepsy neurosurgeons about laterality and localization

of seizure onset based on cognitive test patterns. Though we can make statements about the

sensitivity and specificity of these measures as they relate to frontal and temporal functioning in

general, we cannot decisively comment on their ability to predict seizure localization pre-

surgically. All of the patients in our final sample had medication-refractory epilepsy, though a

portion had undergone resective surgery to alleviate their seizures. Cortical resections for

epilepsy tend to be rather focal removing only the epileptogenic foci if possible, though often

surrounding tissue may also be removed or compromised. For our post-surgical patients, surgical

intervention tended to be curative, as the majority were seizure free following their resections.

This alleviation of seizures likely promoted overall brain health, but could have also introduced









more cortical damage than was initially caused by the seizures themselves. This could have

obscured the true cognitive picture that may have been present in a strictly pre-surgical

population, making conclusions about performance in that population difficult. Unfortunately,

our samples of pre and post-surgical patients were too small to conduct any meaningful analyses

to compare group differences with regard to demographic, medical, or cognitive variables for

these patients.

Finally, our small sample size prohibited us from examining other factors of interest that

may have impacted our fluency and naming results. This includes an analysis of the role of age

of seizure onset, the absence/presence of lesions, propagation of seizure activity, and effects of

anti-epileptic medication.

Directions for Future Research and Clinical Use

Phonemic and semantic fluency measures are two of the most commonly used tests by

neuropsychologists. These tests have well established, demographically corrected norms, making

them appropriate for Caucasians and African-Americans, and persons from their first through

their ninth decade of life (Delis, Kaplan & Kramer, 2001; Heaton, Miller, Taylor & Grant, 2004).

These tests have been used to characterize the cognitive performance of virtually every type of

patient, including those with dementia, epilepsy, TBI, infection, and tumor. They obviously offer

clinical value as part of neuropsychological assessments and will continue to be used in the

future.

The current study suggests that used individually, they may not offer definitive localizing

value with regard to clinical epilepsy patients. As previously mentioned, the sample of epilepsy

patients in this study is thought to be representative of the overall population of patients who

present for evaluation in surgical epilepsy centers. In our sample, scores on tests of phonemic

and semantic fluency were overlapping, and did not distinguish between patient groups, which is









part of the usual question that motivates pre-surgical neuropsychological assessments. As such,

configural interpretation of cognitive profiles should include fluency tests along with other tests

sensitive to the presence of localized neuronal dysfunction.

Action fluency is a relatively new construct in the literature and has only been used in

limited populations at the present time. These populations tend to have both frontal and

subcortical disease involvement, making it difficult to pinpoint the exact underlying process that

is impairing action fluency performance. Furthermore, no definitive theory has emerged in the

literature that is sufficient to explain the neural underpinnings supporting this construct to the

exclusion of other viewpoints. Complicating the theoretical debate is that fact that cognitive,

neuroimaging, and lesion data exist that support both main theories, raising the possibility that

they may not be mutually exclusive. Our findings do not provide significant clarity to this

debate.

To our knowledge, there have been no published studies that have examined the action

fluency construct in a surgical epilepsy population. Based on the current findings, there appears

to be limited support for the use of an action fluency paradigm in clinical epilepsy at the present

time. To the extent that our sample represents a real-life population, the test appears to have little

predictive power and its' sensitivity to frontal-lobe functioning remains questionable. A

significant amount of theory-driven research needs to be done to show the specificity of this test

and its' construct validity before it is used clinically. More specifically, a series of well-designed

studies contrasting patients with circumscribed premotor/supplementary motor and DLPFC

lesions would shed light on the most important cortical regions subserving test performance.

Additional neuroimaging studies employing an action fluency paradigm compared to generation

of proper nouns, for instance, would also help elucidate brain regions critical to action word









generation in particular. Contrasting patients with isolated focal or subcortical disease would

also provide clarification the relative role of these structures in action fluency performance.

Finally, a more controlled study comparing patients with frontal lobe versus temporal lobe

epilepsy would help to elucidate the contribution of these various structures to action fluency

performance.

To the best of our knowledge, this is the first study that has used a proper noun generation

paradigm embedded within a fluency test, as this construct has been studied specifically within a

naming format to this point. The naming literature has offered preliminary support for this

construct, possibly as more unique to anterior temporal functioning than tests employing naming

of common objects. The results that we found are promising and provide converging evidence

that this may indeed be a paradigm, both in the naming and fluency format, that shows great

promise with a variety of clinical populations.

Additional research with this fluency and naming paradigm is absolutely warranted. First,

further psychometric studies need to be undertaken to help establish the construct validity of

these measures and show the convergent and discriminant validity with other tests in both

clinically and normally-functioning populations. In order for the test to be used clinically,

adequate normative studies need to establish performance in healthy controls on the fluency

paradigm. Development of a standardized naming paradigm could prove to be more difficult due

to stimulus selection, which is something that was carefully considered in the present study.

Naming stimuli have to be free of identifying features (i.e., uniforms, for example), stratified

across decade, and controlled for type and amount of fame (i.e., sports, news, politics, movies).

As we found in the present study, familiarity with face stimuli will vary across person and must

be accounted for. This should be then be contrasted with naming ability in order to gain an









accurate appreciation of true word-retrieval deficits. Additional complications with developing a

clinical famous faces naming task include the fact that it must be updated over time to account

for famous persons now obsolete, or new famous persons in popular media.

Nonetheless, the famous name paradigm, particularly the fluency format, should be studied

in additional clinical epilepsy populations to document its utility with this sample. Preliminary

studies could utilize post-ATL populations, patients with circumscribed lesions, to show the

sensitivity and perhaps specificity to the anterior temporal lobe. The paradigm could then be

advanced to pre-surgical populations to ascertain its ability to aid in the differential diagnosis of

refractory seizure populations, the point of the present study. This paradigm could also offer

significant value to other clinical populations apart from epilepsy. Similar to patients with left

temporal lobe epilepsy, patients with Alzheimer's dementia evidence disease of temporal lobe

structures and perform poorly on naming and semantic fluency paradigms. Name fluency may

prove to be sensitive to degenerative disease of the temporal structures as well. Of more interest

are patients with "pre-Alzheimer's", or mild cognitive impairment (MCI). Most of these patients

evidence focal impairment in one domain, but later go on to progress to full blown dementia. As

of late, the focus of research has turned to identifying these patients in a pre-clinical state so that

early intervention may be undertaken. As many of these patients report that their earliest problem

is difficulty retrieving names of people, this measure may provide a useful clinical tool in

identifying early deficits not tapped by other assessment measures.

Finally, our findings with regard to qualitative fluency performance were discordant than

most of those presented in the literature, but may be explained by overall lack of fluency

differences across groups. As this finding of related qualitative-quantitative performance has

been reported elsewhere, it is at least a likely possibility that these scores are dependent. Given









that, the clinical utility of clusters and switches necessarily depends on how sensitive fluency

measures are at detecting the presence or absence of pathology.

Clustering and switching analyses are easy to compute and do not require significant

additional work once familiarity with the scoring criteria has been established, making them

relatively easy measures to consider clinically. However, additional work needs to be done to

establish a normative basis for these measures before they are used in impaired populations. At

the present time, only one study has attempted to norm these measures (Troyer, 2000). Should

normative data be established, these measures could be used clinically in conjunction with

overall fluency score to help provide useful information about localization of function.










APPENDIX A
STANDARD NEUROPSYCHOLOGICAL TEST BATTERY (SNB)


Construct Test Name and Reference Description of Test Measure Dependent Variables;
Time to Administer
Intellectual Wechsler Abbreviated Block Design, Matrix Verbal IQ, Performance IQ,
Functioning Scale of Intelligence Reasoning, Vocabulary, and Full Scale IQ; subtest T-
(WASI; Psychological Similarities subtests scores Time=20-30'
Corporation, 1999)
Memory Functioning Rey Complex Figure Test Measure of figural memory Scores for Immediate and
and Recognition Trial visuoconstruction that requires Delayed Recall
(Myers & Myers, 2004) copy, immediate and delayed Time=15'
recall of a complex figure
California Verbal Verbal list learning; assesses Scores for Immediate and
Learning Test 2nd learning strategy, immediate Delayed Free Recall
Edition (Delis et al., 2000) and delayed recall, Time=15'
recognition, and interference
Wechsler Memory Scale- Measure of verbal memory for Scores for Immediate and
R; Logical Memory I & stories and figural memory for Delayed Recall
III (Wechsler, 1997) geometric designs. Time=15'
Language Controlled Oral Word Verbal fluency for alphabet Total correct exemplars
Functioning Association (COWA; letter (i.e. F,A,S) Time=5'
Spreen & Benton, 1977)
Semantic Fluency Verbal fluency for a semantic Total correct exemplars
Animals (Tombaugh et al., category Time=5'
1999)
Boston Naming Test II Confrontation naming using Total Correct
(Goodglass & Kaplan, large ink drawings Time=10'
2000)
Visuo-perceptual / WASI Block Design Visuoconstructional measure Total score based on time
Visuo-constructional subtest (Wechsler, 1997) requiring construction with limits Time=10'
Functioning blocks
Frontal /Executive Wisconsin Card Sorting Measure of mental flexibility Number of categories
Skills -- Attention, Test (Heaton, 1981) and problem solving, achieved; Number of errors;
Psychomotor Speed, trials to finish first category
Abstract Thinking Time= 15'
Trail Making Test Measures visuomotor speed, Total time to complete;
(Reitan, 1958) set-shifting Number of errors
Total time=5'
WAIS-III Digit Span Memory for digit sequences; Total correct score
(Wechsler, 1997) Requires attention span Time=10'
Sensory Perceptual Finger Tapping (Halstead, Speeded fine motor movement Average number of taps
and Motor 1947; Reitan & Wolfson, for dominant and non- across five trials
1993) dominant hands Time=5'
Grooved Pegboard Test Speeded fine motor movement Total time; # of drops
(Klove, 1963) and dexterity Time=5'
Mood and Affect Beck Depression 21 item self-evaluation Total number of items
Inventory-II (Beck, Steer, questionnaire assessing endorsed
& Brown, 1996 ) elements of depression Time=5'









APPENDIX B
SAMPLE RESPONSES FROM FLUENCY DATA


Presidents


Bill Clinton )
Hillary Clinton
John Kerry
John Edwards Politicians
John McCain
Barack Obama
Beyonce
Johnny Depp
Kirstin Dunst Actors/Ac
Tom Hanks
Meg Ryan
Elizabeth Taylor
Spencer Tracy
atherine Hepburr Tempo
George Burns Grou
Gracie Allen -
George Harrison
John Lennon Musicians
Ringo Starr
aul McCartney
Dakota Fanning
Ricky Martin


:tresses


,rally
ped









Action Fluency


- Feet/legs


4 Exercise


Mouth
Swallow Mouth
Sip
Sleep
Lay | Rest/Sleep
Relax
Rest
Smile
Blink Facial Gestures
Squint
Think
Kick
(Toss
Hurl
Throw Hands
Put
Pedal
Ride
Reel









REFERENCES


Alexander, M. P., Stuss, D. T., Shallice, T., Picton, T. W., & Gillingham, S. (2005). Impaired
concentration due to frontal lobe damage from two distinct lesion sites. Neurology, 65,
572-579.

Alvarez, J. A., & Emory, E. (2006). Executive function and the frontal lobes: A meta-analytic
review. Neuropsychology Reviews, 16, 17-42.

Annegers, J. F., Hauser, W. A., Coan, S. P., & Rocca, W. A. (1998). A population-based study of
seizures after traumatic brain injuries. New England Journal of Medicine, 338, 20-24.

Annegers, J. F., Hauser, W. A., Lee, J. R., & Rocca, W. A. (1995). Secular trends and birth
cohort effects in unprovoked seizures: Rochester, Minnesota 1935-1984. Epilepsia, 36,
575-579.

Baldo, J.V., Schwartz, S., Wilkins, D., & Dronkers, N.F. (2006). Role of frontal versus temporal
cortex in verbal fluency as revealed by voxel-based lesion symptom mapping. Journal of
the International Neuropsychological Society, 12, 896-900.

Baldo, J. V., & Shimamura, A. P. (1998). Letter and category fluency in patients with frontal
lobe lesions. Neuropsychology, 12, 259-267.

Barr, W. B., Goldberg, E., Wasserstein, J., & Novelly, R. A. (1990). Retrograde amnesia
following unilateral temporal lobectomy. Neuropsychologia, 28, 243-255.

Bechara, A., Damasio, H., & Damasio, A. R. (2000). Emotion, decision making and the
orbitofrontal cortex. Cerebral Cortex, 10, 295-307.

Blenner, J.L., (1993). The discriminant capacity of the Stroop test in tumor neurosurgical
patients and its relationship to the visual evoked potential measure. Personality and
Individual Differences, 15, 99-102.

Bredart, S. (1993). Retrieval failures in face naming. Memory, 1, 351-366.

Butler, M., Retzlaff, P.D., & Vanderploeg, R. (1991). Neuropsychological test usage.
Professional Psychology: Research and Practice, 22, 510-512.

Butters, N., Granholm, E., Salmon, D. P., Grant, I., & Wolfe, J. (1987). Episodic and semantic
memory: a comparison of amnesic and demented patients. Journal of Clinical and
Experimental Neuropsychology, 9, 479-497.

Cabeza R., & Nyberg, L.(2000). Neural bases of learning and memory: functional
neuroimaging evidence. Current Opinions in Neurology, 13, 415-421.









Cappa, S. F., Binetti, G., Pezzini, A., Padovani, A., Rozzini, L., & Trabucchi, M. (1998). Object
and action naming in Alzheimer's disease and frontotemporal dementia. Neurology, 50,
351-355.

Caramazza, A., & Hillis, A. E. (1991). Lexical organization of nouns and verbs in the brain.
Nature, 349, 788-790.

Chin, R. F., Neville, B. G., & Scott, R. C. (2005). Meningitis is a common cause of convulsive
status epilepticus with fever. Archives of Disease in Childhood, 90, 66-69.

Corcoran, R., & Upton, D. (1993). A role for the hippocampus in card sorting? Cortex, 29, 293-
304.

Costello, A. L., & Warrington, E. K. (1989). Dynamic aphasia: the selective impairment of
verbal planning. Cortex, 25, 103-114.

Cotelli, M., Borroni, B., Manenti, R., Alberici, A., Calabria, M., Agosti, C., et al. (2006). Action
and object naming in frontotemporal dementia, progressive supranuclear palsy, and
corticobasal degeneration. Neuropsychology, 20, 558-565.

Damasio, A.R. (1995). On some functions of the human prefrontal cortex. Annals of the
New York Academy ofScience, 769, 241-251.

Damasio, H., Grabowski, T. J., Tranel, D., Ponto, L. L., Hichwa, R. D., & Damasio, A. R.
(2001). Neural correlates of naming actions and of naming spatial relations. Neuroimage,
13(6 Pt 1), 1053-1064.

Damasio, A. R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed
neural systems. Proceedings of the NationalAcademies of Science USA, 90, 4957-4960.

Damasio, H., Tranel, D., Grabowski, T., Adolphs, R., & Damasio, A. (2004). Neural systems
behind word and concept retrieval. Cognition, 92, 179-229.

Daniele, A., Giustolisi, L., Silveri, M. C., Colosimo, C., & Gainotti, G. (1994). Evidence for a
possible neuroanatomical basis for lexical processing of nouns and verbs.
Neuropsychologia, 32, 1325-1341.

Davies, K. G., Hermann, B. P., Dohan, F. C., Jr., & Wyler, A. R. (1996). Intractable epilepsy due
to meningitis: results of surgery and pathological findings. British Journal of
Neurosurgery, 10, 567-570.

Delis, D.C., Kaplan, E., and Kramer, J.H. (2001). The Delis-Kaplan Executive Function System.
San Antonio: The Psychological Corporation.









De N6brega E, Nieto A, Barroso J, Mont6n F. (2007). Differential impairment in semantic,
phonemic, and action fluency performance in Friedreich's ataxia: possible evidence of
prefrontal dysfunction. Journal of the International Neuropsychological Society, 13, 944-
52.

Demakis, G. J. (2004). Frontal lobe damage and tests of executive processing: a meta-analysis of
the category test, stroop test, and trail-making test. Journal of Clinical and Experimental
Neuropsychology, 26, 441-450.

Demakis, G. J., Mercury, M. G., Sweet, J. J., Rezak, M., Eller, T., & Vergenz, S. (2003).
Qualitative analysis of verbal fluency before and after unilateral pallidotomy. Clinical
Neuropsychology, 17, 322-330.

Diaz, M., Sailor, K., Cheung, D., & Kuslansky, G. (2004). Category size effects in semantic and
letter fluency in Alzheimer's patients. Brain and Language, 89, 108-114.

Drane, D. L., Lee, G. P., Cech, H., Huthwaite, J. S., Ojemann, G. A., Ojemann, J. G., et al.
(2006). Structured cueing on a semantic fluency task differentiates patients with temporal
versus frontal lobe seizure onset. Epilepsy and Behavior, 9, 339-344.

Drane, D.L., Ojemann, G.A., Aylward, E., Ojemann, J.G., Johnson, L.C., Silbergeld, D.L.,
Miller, J.W., & Tranel, D. (2008). Category-specific naming and recognition deficits in
temporal lobe epilepsy surgical patients. Neuropsychologia, 46, 1242-55.

Exner, C., Boucsein, K., Lange, C., Winter, H., Weniger, G., Steinhoff, B. J., et al. (2002).
Neuropsychological performance in frontal lobe epilepsy. Seizure, 11, 20-32.

Forsgren, L., Beghi, E., Oun, A., & Sillanpaa, M. (2005). The epidemiology of epilepsy in
Europe a systematic review. European Journal ofNeurology, 12, 245-253.

Fossati, P., Guillaume le, B., Ergis, A. M., & Allilaire, J. F. (2003). Qualitative analysis of verbal
fluency in depression. Psychiatry Research, 117, 17-24.

Frith, C. D., Friston, K. J., Herold, S., Silbersweig, D., Fletcher, P., Cahill, C., et al. (1995).
Regional brain activity in chronic schizophrenic patients during the performance of a
verbal fluency task. British Journal of Psychiatry, 167, 343-349.

Frith, C. D., Friston, K. J., Liddle, P. F., & Frackowiak, R. S. (1991). A PET study of word
finding. Neuropsychologia, 29, 1137-1148.

Fukatsu, R., Fujii, T., Tsukiura, T., Yamadori, A., & Otsuki, T. (1999). Proper name anomia
after left temporal lobectomy: a patient study. Neurology, 52, 1096-1099.

Gleissner, U., & Elger, C. E. (2001). The hippocampal contribution to verbal fluency in patients
with temporal lobe epilepsy. Cortex, 37, 55-63.









Glosser, G., & Donofrio, N. (2001). Differences between nouns and verbs after anterior temporal
lobectomy. Neuropsychology, 15, 39-47.

Glosser, G., Salvucci, A. E., & Chiaravalloti, N. D. (2003). Naming and recognizing famous
faces in temporal lobe epilepsy. Neurology, 61, 81-86.

Goldberg, E & Bougakov, D. (2005). Neuropsychologic assessment of frontal lobe dysfunction.
Psychiatric Clinics of North America, 28, 567-580.

Gomo-Tempini, M. L., Price, C. J., Josephs, O., Vandenberghe, R., Cappa, S. F., Kapur, N., et
al. (1998). The neural systems sustaining face and proper-name processing. Brain, 121
(Pt 11), 2103-2118.

Gourovitch, M. L., Kirkby, B. S., Goldberg, T. E., Weinberger, D. R., Gold, J. M., Esposito, G.,
et al. (2000). A comparison ofrCBF patterns during letter and semantic fluency.
Neuropsychology, 14, 353-360.

Grabowski, T. J., Damasio, H., Tranel, D., Ponto, L. L., Hichwa, R. D., & Damasio, A. R.
(2001). A role for left temporal pole in the retrieval of words for unique entities. Human
Brain Mapping, 13, 199-212.

Harris, D.M., & Kay, J. (1995). I recognize your face but can't remember your name: Is it
because names are unique? British Journal of Psychology, 86, 345-58.

Heaton, R.K., Miller, S.W., Taylor, M.J., & Grant, I. (2004) Revised Comprehensive Norms for
an Expanded Halstead-Reitan Battery. Psychological Assessment Resources, Odessa,
FL.

Helmstaedter, C. (2001). Behavioral aspects of frontal lobe epilepsy. Epilepsy and Behavior, 2,
384-395.

Helmstaedter, C., Kemper, B., & Elger, C. E. (1996). Neuropsychological aspects of frontal lobe
epilepsy. Neuropsychologia, 34, 399-406.

Henry, J. D., & Crawford, J. R. (2004a). A meta-analytic review of verbal fluency performance
in patients with traumatic brain injury. Neuropsychology, 18, 621-628.

Henry, J. D., & Crawford, J. R. (2004b). A meta-analytic review of verbal fluency performance
following focal cortical lesions. Neuropsychology, 18, 284-295.

Hermann, B.P., Perrine, K., Chelune, G.J., Barr, W., Loring, D.W., Strauss, E., Trenerry,
M.R., & Westerveld, M. (1999). Visual confrontation naming following left
anterior temporal lobectomy: a comparison of surgical approaches.
Neuropsychology, 13, 3-9.









Hermann, B., & Seidenberg, M. (1995). Executive system dysfunction in temporal lobe epilepsy:
effects of nociferous cortex versus hippocampal pathology. Journal of Clinical and
Experimental Neuropsychology, 17, 809-819.

Hermann, B., Seidenberg, M., Bell, B., Rutecki, P., Sheth, R.D., Wendt, G., O'Leary, D.,
& Magnotta, V. (2003) Extratemporal quantitative MR volumetrics and
neuropsychological status in temporal lobe epilepsy. Journal of the International
Neuropsychological Society, 9, 353-362.

Hermann, B. P., Seidenberg, M., Dohan, F. C., Jr., Wyler, A. R., Haltiner, A., Bobholz, J., et al.
(1995). Reports by patients and their families of memory change after left anterior
temporal lobectomy: relationship to degree of hippocampal sclerosis. Neurosurgery, 36,
39-44; discussion 44-35.

Hermann, B. P., Seidenberg, M., Haltiner, A., & Wyler, A. R. (1995). Relationship of age at
onset, chronologic age, and adequacy of preoperative performance to verbal memory
change after anterior temporal lobectomy. Epilepsia, 36, 137-145.

Hermann, B. P., Seidenberg, M., Schoenfeld, J., & Davies, K. (1997). Neuropsychological
characteristics of the syndrome of mesial temporal lobe epilepsy. Archives ofNeurology,
54, 369-376.

Hermann, B. P., Wyler, A. R., & Richey, E. T. (1988). Wisconsin Card Sorting Test performance
in patients with complex partial seizures of temporal-lobe origin. Journal of Clinical and
Experimental Neuropsychology, 10, 467-476.

Ho, A.K., Sahakia, B.J., Robbins, T.W., Barker, R.A., Rosser, A.E., & Hodges, J.R. (2002).
Verbal fluency in Huntington's disease: A longitudinal analysis on phonemic and
semantic clustering and switching. Neuropsychologia, 40, 1277-1284.

Janowsky, J. S., Shimamura, A. P., & Squire, L. R. (1989). Source memory impairment in
patients with frontal lobe lesions. Neuropsychologia, 27, 1043-1056.

Joanette, Y., & Goulet, P. (1986). Criterion-specific reduction of verbal fluency in right brain-
damaged right-handers. Neuropsychologia, 24, 875-879.

Jokeit, H., & Ebner, A. (2002). Effects of chronic epilepsy on intellectual functions. Progress in
Brain Research, 135, 455-463.

Lamar, M. & Resnick, S.M. (2004). Aging and prefrontal functions: dissociating orbitofrontal
and dorsolateral abilities. Neurobiology ofAging. 25, 553-338.

Lezak, M.D., Howieson, D.B. & Loring, D.W. (2004). NeuropsychologicalAssessment
(4 ed). New York: Oxford University Press.









Loring, D.W., Marino, S., & Meador, K.J. (2007). Neuropsychological and behavioral
effects of antiepilepsy drugs. Neuropsychological Review, 17, 413-425.

Lu, L. H., Crosson, B., Nadeau, S. E., Heilman, K. M., Gonzalez-Rothi, L. J., Raymer, A., et al.
(2002). Category-specific naming deficits for objects and actions: semantic attribute and
grammatical role hypotheses. Neuropsychologia, 40, 1608-1621.

Luria, A.R. (1966). Higher cortical functions in man. New York: Basic Books.

Martin, A., & Fedio, P. (1983). Word production and comprehension in Alzheimer's disease: the
breakdown of semantic knowledge. Brain and Language, 19, 124-141.

Martin, R. C., Loring, D. W., Meador, K. J., & Lee, G. P. (1990). The effects of lateralized
temporal lobe dysfunction on formal and semantic word fluency. Neuropsychologia, 28,
823-829.

Martin, R.C., Sawrie, S.M., Gilliam, F.G., Palmer, C.A., Faught, E., Morawetz, R.B., &
Kuzniecky, R.I. (2000). Wisconsin Card Sorting performance in patients with temporal
lobe epilepsy: clinical and neuroanatomical correlates. Epilepsia, 41, 1626-1632.

Martins, I. P., & Farrajota, L. (2007). Proper and common names: a double dissociation.
Neuropsychologia, 45, 1744-1756.

McCarthy, R., & Warrington, E. K. (1985). Category specificity in an agrammatic patient: the
relative impairment of verb retrieval and comprehension. Neuropsychologia, 23, 709-
727.

McDonald, C. R., Bauer, R. M., Filoteo, J. V., Grande, L., Roper, S. N., & Gilmore, R. (2006).
Episodic memory in patients with focal frontal lobe lesions. Cortex, 42, 1080-1092.

McDonald, C. R., Delis, D. C., Norman, M. A., Tecoma, E. S., & Iragui, V. J. (2005).
Discriminating patients with frontal-lobe epilepsy and temporal-lobe epilepsy: utility of a
multilevel design fluency test. Neuropsychology, 19, 806-813.

McDonald, C. R., Delis, D. C., Norman, M. A., Tecoma, E. S., & Iragui-Madozi, V. I. (2005). Is
impairment in set-shifting specific to frontal-lobe dysfunction? Evidence from patients
with frontal-lobe or temporal-lobe epilepsy. Journal of the International
Neuropsychological Society, 11, 477-481.

McDonald, C. R., Delis, D. C., Norman, M. A., Wetter, S. R., Tecoma, E. S., & Iragui, V. J.
(2005). Response inhibition and set shifting in patients with frontal lobe epilepsy or
temporal lobe epilepsy. Epilepsy and Behavior, 7, 438-446.

McKenna, P., & Warrington, E.K. (1980). Testing for nominal dysphasia. Journal ofNeurology,
Neurosurgery, andPsychiatry, 43, 781-788.









McMillan, A.B., Hermann, B.P., Johnson, S.C., Hansen, R.R., Seidenberg, M., & Meyerand,
M.E. (2004). Voxel-based morphometry of unilateral temporal lobe epilepsy reveals
abnormalities in cerebral white matter. Neuroimage, 23, 167-174.

Miceli, G., Silveri, M. C., Villa, G., & Caramazza, A. (1984). On the basis for the agrammatic's
difficulty in producing main verbs. Cortex, 20, 207-220.

Milders, M. (2000). Naming famous faces and buildings. Cortex, 36, 138-145.

Miller, E. (1984). Verbal fluency as a function of a measure of verbal intelligence and in relation
to different types of cerebral pathology. British Journal of Clinical Psychology, 23 (Pt
1), 53-57.

Monsch, A. U., Bondi, M. W., Butters, N., Salmon, D. P., Katzman, R., & Thal, L. J. (1992).
Comparisons of verbal fluency tasks in the detection of dementia of the Alzheimer type.
Archives ofNeurology, 49, 1253-1258.

Mummery, C. J., Patterson, K., Hodges, J. R., & Wise, R. J. (1996). Generating 'tiger' as an
animal name or a word beginning with T: differences in brain activation. Proceedings:
Biological Science, 263, 989-995.

N'Kaoua, B., Lespinet, V., Barsse, A., Rougier, A., & Claverie, B. (2001). Exploration of
hemispheric specialization and lexico-semantic processing in unilateral temporal lobe
epilepsy with verbal fluency tasks. Neuropsychologia, 39, 635-642.

Nichols, M.E., Meador, K.J., & Loring, D.W. (1993). Neuropsychological effects of
antiepileptic drugs: a current perspective. Clinical Neuropharmacology, 16,
471-484

Obler, L.K., & Albert, M.L. (1979). The Action Naming Test (Experimental ed.) Boston: VA
Medical Center.

Oldfield, R.C. (1971). The assessment and analysis of handedness: the Edinburgh inventory.
Neuropsychologia, 9, 97-113.

Ostberg, P., Fernaeusm S.E., Hellstromm K., Bogdanovicm N., & Wahlundm L,O.
(2005). Impaired verb fluency: a sign of mild cognitive impairment. Brain and
Language, 95, 273-279.

Owen, A., Downes, J., Sahakian, B., Polkey, C., & Robbins, T. (1990). Planning and spatial
working memory following frontal lobe lesions in man. Neuropsychologia, 28, 1021-
1034.

Oyegbile, T.O., Bhattacharya, A., Seidenberg, M., & Hermann, B.P. (2006). Quantitative MRI
biomarkers of cognitive morbidity in temporal lobe epilepsy. Epilepsia, 47, 143-152.









Paradowski, B., & Zagrajek, M. M. (2005). Epilepsy in middle-aged and elderly people: a three-
year observation. Epileptic Disorders, 7, 91-95.

Parks, R.W., Loewenstein, D.A., Dodrill, K.L., Barker, W.W., Yoshii, F., Chang, J.Y., Emran,
A., Apicella, A., Sheramata, W.A., & Duara, R. (1988). Cerebral metabolic effects of a
verbal fluency test: A PET scan study. Journal of Clinical and Experimental
Neuropsychology, 10, 565-575.

Paulesu, E., Goldacre, B., Scifo, P., Cappa, S. F., Gilardi, M. C., Castiglioni, I., et al. (1997).
Functional heterogeneity of left inferior frontal cortex as revealed by fMRI. Neuroreport,
8,2011-2017.

Pendleton, M. G., Heaton, R. K., Lehman, R. A., & Hulihan, D. (1982). Diagnostic utility of the
Thurstone Word Fluency Test in neuropsychological evaluations. Journal of Clinical
Neuropsychology, 4, 307-317.

Peran, P., Cardebat, D., Cherubini, A., Piras, F., Luccichenti, G., Peppe, A., Caltagirone, C.,
Rascol, O., Demonet, J.F., & Sabatini, U. (2009). Object naming and action-verb
generation in Parkinson's disease: A fMRI study. Cortex [Epub ahead ofprint]. Available
online March 14, 2009.

Perani, D., Cappa, S. F., Schnur, T., Tettamanti, M., Collina, S., Rosa, M. M., et al. (1999). The
neural correlates of verb and noun processing. A PET study. Brain, 122 (Pt 12), 2337-
2344.

Perret, E. (1974). The left frontal lobe of man and the suppression of habitual responses in verbal
categorical behaviour. Neuropsychologia, 12, 323-330.

Perucca, E. (2005). An introduction to antiepileptic drugs. Epilepsia, 46 Suppl 4, 31-37.

Petersen, S.E., Fox, P.T., Posner, M.I., Mintum, M.A., & Raichle, M.A. (1989). Positron
emission tomographic studies of the processing of single words. Journal of Cognitive
Neuroscience, 1, 153-170.

Petersen, S. E., Fox, P. T., Snyder, A. Z. & Raichle, M. E. (1990). Activation of extrastriate and
frontal cortical areas by visual words and word-like stimuli.
Science, 249, 1041-1044.

Petrides, M. & Milner, B. (1982). Deficits on subject-ordered tasks after frontal- and temporal-
lobe lesions in man. Neuropsychologia, 20, 249-62.

Piatt, A. L., Fields, J. A., Paolo, A. M., Koller, W. C., & Troster, A. I. (1999). Lexical, semantic,
and action verbal fluency in Parkinson's disease with and without dementia. Journal of
Clinical and Experimental Neuropsychology, 21, 435-443.









Piatt, A. L., Fields, J. A., Paolo, A. M., & Troster, A. I. (1999). Action (verb naming) fluency as
an executive function measure: convergent and divergent evidence of validity.
Neuropsychologia, 37, 1499-1503.

Raichle, M. E., Fiez, J. A., Videen, T. O., MacLeod, A. M., Pardo,J. V., Fox, P. T. & Petersen, S.
E. (1994) Practice-related changes in human brain functional anatomy during non-motor
learning. Cerebral Cortex 4, 8-26.

Ramier, A. M., & Hecaen, H. (1970). [Respective roles of frontal lesions and lesion lateralization
in "verbal fluency" deficiencies]. Rev Neurol (Paris), 123(1), 17-22.

Randolph, C., Braun, A.R., Goldberg, T.E., & Chase, T.N. (1993). Semantic fluency in
Alzheimer's, Parkinson's, and Huntington's Disease: Dissociation of Storage and
Retrieval Failures. Neuropsychology, 7, 82-88.

Rapp, B., & Caramazza, A. (1998). A case of selective difficulty in writing verbs. Neurocase, 4,
127-139.

Rende, B., Ramsberger, G., & Miyake, A. (2002). Commonalities and differences in the working
memory components underlying letter and category fluency tasks: a dual-task
investigation. Neuropsychology, 16, 309-321.

Reverberi, C., Laiacona, M., & Capitani, E. (2006). Qualitative features of semantic fluency
performance in mesial and lateral frontal patients. Neuropsychologia, 44, 469-478.

Risberg, J., & Grafman, J. (2006). The frontal lobes: Development, function, and pathology.
Cambridge, UK: Cambridge University Press.

Robert, P.H., Lafont, V., Medecin, I., Berthet, L., Thauby, S., Baudu, C., & Darcourt, G. (1998).
Clustering and switching in verbal fluency tasks: Comparison between schizophrenics
and healthy adults. Journal of the International Neuropsychological Society, 4, 539-546.

Rosser, A., & Hodges, J. R. (1994). Initial letter and semantic category fluency in Alzheimer's
disease, Huntington's disease, and progressive supranuclear palsy. Journal ofNeurology,
Neurosurgery, and Psychiatry, 57, 1389-1394.

Ruff, R. M., Light, R. H., Parker, S. B., & Levin, H. S. (1997). The psychological construct of
word fluency. Brain and Language, 57, 394-405.

Seidenberg, M., Griffith, R., Sabsevitz, D., Moran, M., Haltiner, A., Bell, B., et al. (2002).
Recognition and identification of famous faces in patients with unilateral temporal lobe
epilepsy. Neuropsychologia, 40, 446-456.

Semenza, C., & Zettin, M. (1989). Evidence from aphasia for the role of proper names as pure
referring expressions. Nature, 342, 678-679.









Shapiro, K. A., Moo, L. R., & Caramazza, A. (2006). Cortical signatures of noun and verb
production. Proceedings of the National Academies of Science USA, 103, 1644-1649.

Silveri, M. C., Salvigni, B. L., Cappa, A., Della Vedova, C., & Puopolo, M. (2003). Impairment
of verb processing in frontal variant-frontotemporal dementia: a dysexecutive symptom.
Dementia and Geriatric Cognitive Disorders, 16, 296-300.

Sperling, M. R. (2004). The consequences of uncontrolled epilepsy. CNS Spectrums, 9(2), 98-
101.

Spreen, O. & Strauss, E. (1998). A Compendium ofNeuropsychological Tests. New York:
Oxford University Press.

Stuss, D. T., & Alexander, M. P. (2000). Executive functions and the frontal lobes: a conceptual
view. Psychological Research, 63(3-4), 289-298.

Stuss, D. T., Alexander, M. P., Hamer, L., Palumbo, C., Dempster, R., Binns, M., et al. (1998).
The effects of focal anterior and posterior brain lesions on verbal fluency. Journal of the
International Neuropsychological Society, 4, 265-278.

Stuss, D. T., Floden, D., Alexander, M. P., Levine, B., & Katz, D. (2001). Stroop performance in
focal lesion patients: dissociation of processes and frontal lobe lesion location.
Neuropsychologia, 39, 771-786.

Stuss, D. T., & Levine, B. (2002). Adult clinical neuropsychology: lessons from studies of the
frontal lobes. Annual Reviews of Psychology, 53, 401-433.

Stuss, D. T., Toth, J. P., Franchi, D., Alexander, M. P., Tipper, S., & Craik, F. I. (1999).
Dissociation of attentional processes in patients with focal frontal and posterior lesions.
Neuropsychologia, 37, 1005-1027.

Thaiss, L. & Petrides, M. (2003), Source versus content memory in patients with a unilateral
frontal cortex or a temporal lobe excision. Brain, 126, 1112-1126.

Thompson-Schill, S. L., Swick, D., Farah, M. J., D'Esposito, M., Kan, I. P., & Knight, R. T.
(1998). Verb generation in patients with focal frontal lesions: a neuropsychological test
of neuroimaging findings. Proceedings of the National Academies of Science USA, 95,
15855-15860.

Tranel, D. (1992). Neuropsychological assessment. Psychiatric Clinics of North America, 15,
283-99.

Tranel, D. (2006). Impaired naming of unique landmarks is associated with left temporal polar
damage. Neuropsychology, 20, 1-10.









Tranel, D., Damasio, H., & Damasio, A.R. (1997). A neural basis for the retrieval of
conceptual knowledge. Neuropsychologia, 35, 1319-1327.

Tranel, D., Damasio, H., Eichhorn, G.R., Grabowski, T., Ponto, L.L., & Hichwa, R.D.
(2003). Neural correlates of naming animals from their characteristic sounds.
Neuropsychologia, 41, 847-854.

Tranel, D., Grabowski, T. J., Lyon, J., & Damasio, H. (2005). Naming the same entities from
visual or from auditory stimulation engages similar regions of left inferotemporal
cortices. Journal of Cognitive Neuroscience, 17, 1293-1305.

Trenerry, M., & Jack, C.R. Jr. (1994). Wisconsin Card Sorting Test performance before
and after temporal lobectomy. Journal ofEpilepsy, 7, 313-317.

Troster, A. I., Salmon, D. P., McCullough, D., & Butters, N. (1989). A comparison of the
category fluency deficits associated with Alzheimer's and Huntington's disease. Brain
and Language, 37, 500-513.

Troster, A. I., Warmflash, V., Osorio, I., Paolo, A. M., Alexander, L. J., & Barr, W. B. (1995).
The roles of semantic networks and search efficiency in verbal fluency performance in
intractable temporal lobe epilepsy. Epilepsy Research, 21, 19-26.

Troyer, A.K. (2000). Normative data for clustering and switching on verbal fluency tasks.
Journal of Clinical and Experimental Neuropsychology, 22, 370-378.

Troyer, A. K., & Moscovitch, M. (2006). Cognitive processes of verbal fluency tasks. In Poreh,
Amir M. (ed), The Quantified Process Approach to Neuropsychological Assessment (pp.
143-160). New York: Taylor & Francis.

Troyer, A. K., Moscovitch, M., & Winocur, G. (1997). Clustering and switching as two
components of verbal fluency: evidence from younger and older healthy adults.
Neuropsychology, 11, 138-146.

Troyer, A. K., Moscovitch, M., Winocur, G., Alexander, M. P., & Stuss, D. (1998a). Clustering
and switching on verbal fluency: the effects of focal frontal- and temporal-lobe lesions.
Neuropsychologia, 36, 499-504.

Troyer, A. K., Moscovitch, M., Winocur, G., Leach, L., & Freedman, M. (1998a). Clustering and
switching on verbal fluency tests in Alzheimer's and Parkinson's disease. Journal of the
International Neuropsychological Society, 4, 137-143.

Tsukiura, T., Fujii, T., Fukatsu, R., Otsuki, T., Okuda, J., Umetsu, A., et al. (2002). Neural basis
of the retrieval of people's names: evidence from brain-damaged patients and fMRI.
Journal of Cognitive Neuroscience, 14, 922-937.









Tyler, L.K., Bright, P., Flecther, P., & Stamatakis, E.A. (2004). Neural processing of nouns and
verbs: the role of inflectional morphology. Neuropsychologia, 42, 512-523.

Vendrell, P., Junque, C., Pujol, J., Jurado, M. A., Molet, J., & Grafman, J. (1995). The role of
prefrontal regions in the Stroop task. Neuropsychologia, 33, 341-352.

Vilkki, J., & Holst, P. (1994). Speed and flexibility on word fluency tasks after focal brain
lesions. Neuropsychologia, 32, 1257-1262.

Wakamoto, H., Hayashi, M., Nagao, H., Morimoto, T., & Kida, K. (2004). Clinical investigation
of genetic contributions to childhood-onset epilepsies and epileptic syndromes. Brain &
Development, 26, 184-189.

Warrington, E. K. (2000). Homophone meaning generation: a new test of verbal switching for
the detection of frontal lobe dysfunction. Journal of the International
Neuropsychological Society, 6, 643-648.

Warrington, E. K., & Shallice, T. (1984). Category specific semantic impairments. Brain, 107
(Pt 3), 829-854.

White-Devine, T., Grossman, M., & Robinson, K.M. (1996). Verb confrontation naming and
word-picture matching in Alzheimer's disease. Neuropsychology, 10, 495-503.

Woods, S. P., Carey, C. L., Troster, A. I., & Grant, I. (2005). Action (verb) generation in HIV-1
infection. Neuropsychologia, 43, 1144-1151.









BIOGRAPHICAL SKETCH

Bonnie C. Sachs received a bachelor's degree in psychology from Virginia Tech and a

master's degree in behavioral neuroscience from American University. She worked for two years

as a research assistant at the National Institutes of Health prior to entering the doctoral program

in the Department of Clinical and Health Psychology at the University of Florida. Bonnie earned

her master's degree in clinical psychology from the University of Florida in 2005, and completed

her clinical internship at the Department of Rehabilitation Medicine at Emory University during

the 2008-2009 academic year. Bonnie received her doctoral degree in clinical psychology

(neuropsychology track) from the University of Florida in 2009. Currently, she is employed as a

postdoctoral fellow at the Mayo Clinic. Her main research interests include the neuropsychology

of epilepsy, patterns of cognitive impairment in dementia, neurorehabilitation, and functional

neuroimaging.





PAGE 1

1 THE DIAGNOSTIC UTILITY OF A MULTI TASK VERBAL FLUENCY PARADIGM IN FRONTAL AND TEMPORAL LOBE EPILEPSY: AN ANALYSIS OF FLUENCY TYPE AND QUALITATIVE PERFORMANCE By BONNIE COLLEEN SACHS 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 2009

PAGE 2

2 2009 Bonnie Colleen Sachs

PAGE 3

3 To my chair, Dr. Russell M. Bauer, and all of my other mentors who have helped me naviga te the world of neuropsychology

PAGE 4

4 ACKNOWLEDGMENTS I thank my family for their support and encouragement during graduate school and internship. I also thank the members of my disser tation committee: Russell Bauer, Ph.D.; David Janicke, Ph.D.; David Loring, Ph.D.; and Steven Roper, M.D.; for their thoughtful contributions to this project.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 7 LIST OF FIGURES .............................................................................................................................. 8 ABSTRACT .......................................................................................................................................... 9 CHAPTER 1 BACKGROUND AND SIGNIFICANCE ................................................................................. 11 Frontal and Temporal Lobe Epilepsy ......................................................................................... 11 Measurement of Neuropsyc hological Functioning In FLE and TLE ...................................... 14 Verbal Fluency ..................................................................................................................... 17 Semantic Fluency ................................................................................................................. 18 Phonemic Fluency ............................................................................................................... 20 Novel Techniques for Dissociating Frontal and Temporal Lobe Impairment ........................ 23 Action/Verb Retrieva l ......................................................................................................... 23 Proper Name Retrieval ........................................................................................................ 28 Qualitative Analysis ............................................................................................................ 30 Summary ...................................................................................................................................... 33 2 SPECIFIC AIMS & HYPOTHESES OF CURRENT STUDY ............................................... 35 Aim 1 ............................................................................................................................................ 35 Aim 2 ............................................................................................................................................ 35 Aim 3 ............................................................................................................................................ 36 Aim 4 ............................................................................................................................................ 36 3 SUBJECTS & METHODS ......................................................................................................... 37 Study Participants ........................................................................................................................ 37 Epilepsy Patients .................................................................................................................. 39 Pre -surgical patients ..................................................................................................... 39 Post -surgical patients ................................................................................................... 41 Healthy controls ................................................................................................................... 42 Measures ...................................................................................................................................... 43 Pre -surgical Patients ............................................................................................................ 43 Additional Measures ............................................................................................................ 46 Post -surgical Patients and Healthy C ontrols ...................................................................... 48

PAGE 6

6 4 RESULTS .................................................................................................................................... 51 Aim 1 ............................................................................................................................................ 51 Aim 2 ............................................................................................................................................ 53 Aim 3 ............................................................................................................................................ 54 Aim 4 ............................................................................................................................................ 56 Additional Study Aims ............................................................................................................... 58 5 DISCUSSION .............................................................................................................................. 67 Summary of Findings .................................................................................................................. 67 Interpretation of Findings ........................................................................................................... 72 Semantic and Phonemic Fluency ........................................................................................ 7 2 Action Fluency ..................................................................................................................... 76 Name Fluency ...................................................................................................................... 80 Qualitative Analysis of Fluency Performance ................................................................... 83 Limitations of the Present Study ................................................................................................ 86 Directions for Future Research and Clinical Use ...................................................................... 89 APPENDIX A STANDARD NEUROPSYCHOLOGICAL TEST BATTERY (SNB) .................................. 94 B SAMPLE RESPONSES FROM FLUEN CY DATA ................................................................ 95 REFERENCES ................................................................................................................................... 97 BIOGRAPHICAL SKETCH ........................................................................................................... 109

PAGE 7

7 LIST OF TABLES Table page 3 1 Patient seizure characteristics ................................................................................................ 49 3 2 Demographic and clinical characteristics ............................................................................. 50 4 1 Performance on fluency measures ........................................................................................ 60 4 2 Multivariate analysis of fluency performance in patients .................................................... 60 4 3 Four fl uencies predicting patient group membership .......................................................... 60 4 4 Semantic and name fluencies predicting patient group membership .................................. 60 4 5 Correl ations coefficients for fluency measures (patients only) ........................................... 60 4 6 Correlations coefficients for fluency measures (patients and controls) .............................. 61 4 7 Correlations amongst neuropsychological measures (patients only) .................................. 61 4 8 Correlations amongst neuropsychological measures (patients & controls) ........................ 62 4 9 Performance on naming measures ........................................................................................ 62 4 10 Correlations between naming and fluency measures (patients and controls) ..................... 62 4 11 Correlations between naming and fluency measures (patients only) .................................. 63

PAGE 8

8 LIST OF FIGURES Figure page 4 1 Overall fluency performance across groups ......................................................................... 63 4 2 Receiver operating characteristic (ROC) curve for semantic and name fluencies predicting patient group ......................................................................................................... 64 4 3 Mean number of clusters by group........................................................................................ 65 4 4 Mean number of switches by group ...................................................................................... 65 4 5 Mean cluster size by group .................................................................................................... 66

PAGE 9

9 Abstra ct 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 THE DIAGNOSTIC UTILITY OF A MULTI TASK VERBAL FLUENCY PARADIGM IN FRONTAL AN D TEMPORAL LOBE EPILEPSY: AN ANALYSIS OF FLUENCY TYPE AND QUALITATIVE PERFORMANCE By Bonnie Colleen Sachs August 2009 Chair: Name: Russell M. Bauer Major: Psychology Epilepsy is a chronic, disabling condition affecting roughly 50 out of every 100,000 Americans The most common site for seizure onset is in either the frontal (FLE) or temporal lobes (TLE) of the brain. Identification of the onset of seizures is important, and partly determined by cognitive test scores. Elucidating a pattern of neuropsych ological test performance in FLE and TLE is complicated, but allows for more accurate identification of seizure onset, which is essential in treatment planning. Presently, these cognitive patterns are poorly defined and overlapping, partly due to the lack of specificity of our cognitive tests. The purpose of this study was to evaluate the utility of multiple measures of verbal fluency in the differential diagnosis of intractable FL and TL epilepsy. Patients in the study included pre and post -surgical ref ractory epilepsy patients with either temporal (N=14) or frontal lobe epilepsy (N=7) and healthy age and education matched controls (N=20) Patients completed a battery of neur opsychological tests thought to be sensitive to FL and TL functioning including standard semantic and phonemic fluencies and novel action and name fluencies.

PAGE 10

10 We found significant differences between patients and controls for action and name fluency. Patient groups were not statistically different on fluency measures, though effect s izes indicated FL patients outperformed TL patients on name fluency. A qualitative analysis of fluency (clusters and switches) only differed for patient groups on name fluency as well. Only name and semantic fluency were adequate predictors of patient grou p membership. We found support for the notion that all fluency measures were related to overall intellectual ability and verbal / semantic factors. However, measures of semantic and name fluency were more related to semantic abilities and phonemic and action fluency were also related to measures of executive functioning. Results of the study indicate that in a mixed pre and post -surgical epilepsy population, phonemic, action, and semantic fluency were not specific to frontal and temporal lobe functioning. F urther, qualitative assessments of fluency did not offer significant information about seizur e foci. Name fluency differentiat ed well between patient groups and appears to be a novel measure sensitive to the integrity of the left temporal lobe.

PAGE 11

11 CHAPTER 1 BACKGROUND AND SIGNI FICANCE The purpose of this study was to evaluate the utility of multiple measures of verbal fluency in the differential diagnosis of intractable epilepsy of temporal versus frontal lobe origin. The following sections in the present chapter provide background on the diseases of frontal and temporal lobe epilepsy, and the types of cognitive and neuropsychological deficits that are most common in these populations. The standard neuropsychological assessments used to detect these impai rments are also reviewed, including the limitations of these measures in patients with focal epilepsy. Subsequently, the literature positing material -specific impairments in the production of action words and proper names in patients with frontal and t emporal lobe disease is reviewed in order to substantiate the incorporation of these features into traditional neuropsychological assessment. Finally, specific aims and hypotheses for this study are presented. Frontal and Temporal Lobe Epilepsy Epilepsy is a chronic often disabling disorder affecting between 30 and 60 persons per 100,000 in the United States (Hauser, Annegers, Rocca, 1996). In other countries, prevalence ranges between 30 and 100 cases per 100,000 individuals (Forsgren, Beghi, Oun, & Silla npaa, 2005). While epilepsy accounts for only 5 10% of all disabilities in the United States, the consequences of the disorder can be quite severe, including impairments in quality of life, inability to drive and associated loss of independence, less frequ ent social interaction and lower marriage rates (Sperling, 2004) Further, uncontrolled seizures can lead to neuronal death and physiological dysfunction While many individuals with epilepsy are diagnosed at a young age, epilepsy can begin at any point in the lifespan (Paradowski & Zagradjek, 2005; Hauser, Annegers, Rocca, 1996). Often, the etiology of the disorder is unknown. In some cases, however,

PAGE 12

12 its development has been linked to neoplasm, head injury, family history of seizures, or severe illnesses such as meningitis or encephalitis (Chin, Neville, & Scott, 2005; Wakamoto, et al., 2004; Annegers, Hauser, Coan, et al., 1998; Davies, Hermann, Dohan & Wyler, 1996). In recent years, there have been great advances in treatment options for individuals su ffering from this disorder. Currently, there are dozens of pharmacological agents that can be used to prevent seizures, or to help lessen the frequency or intensity of the epileptic events (see Perucca, 2005 for a review). While drug therapies are often ef fective in managing epilepsy, many cases remain refractory to pharmacologic interventions. Individuals with medication refractory epilepsy, many of whom have complex partial (or focal) epilepsy, often suffer from years of debilitating and dangerous seizure s. The most common types of focal epilepsies arise from either the temporal lobe or frontal lobe (Engel, 1996). Despite the relative number of patients with frontal lobe epilepsy (FLE) temporal lobe epilepsy (TLE) patients may become candidates for seizu re surgery more frequently due to the difficulty in diagnosing and localizing frontal lobe epilepsy. This challenge is accounted for by several different reasons: FLE may be associated with diverse seizure semiologies and EEG recordings in FLE often show w idespread epileptic activity, and neuropsychological profiles of patients with suspected FLE are often not distinct from patients with other epilepsies, including the most common variety, TLE (Exner et al., 2002; Hermann, Wyler, & Rich ey 1988; Helmstaedte r, 1996). Although no signature pattern of impairment exist s and much controversy exists around the executive functions of the frontal lobes, behavioral and cognitive cha racteristics of frontal lobe epilepsy, and frontal lobe damage in general, have bee n described in the literature. While executive functions are complex and multi -faceted, most of the literature concerning executive

PAGE 13

13 dysfunction does implicate frontal lobe damage as etiologic The heterogeneous term, executive function refers to both the notion of cognitive control over mental abilities, and the ability to be adaptive and flexible in novel or unpredictable situations, which is clearly important for everyday functioning. Generally, executive functions can be thought of as behaviors that include problem solving, planning, abstraction, response inhibition, self awareness, cognitive flexibility, cognitive control and hypothesis generation ( Risberg & Grafman, 2006; Lezak et al., 2004). Patients with frontal damage can exhibit various types of e xecutive dysfunction on neuropsychological tests, depending on which sector of the frontal lobe is involved. Deficits include impairments in planning, initiative, inhibition, behavioral control, emotional regulation and working memory (Helmstaedter, 2001) In addition to cognitive deficits, patients with lesions to the frontal lobe can exhibit a variety of characteristic personality, behavioral, and emotional changes. Damage to the orbitofrontal region for instance, can cause behavioral disinhibition, em otional lability, impulsivity, altered social conduct (so -called aquired sociopathy) and changes in personality, whereas damage to the lateral prefrontal cortex can cause a decline in working memory abilities, impairments in abstract reasoning, mental inflexibility, and difficulties with decision making (Bechara, Damasio, & Damasio, 2000; Tranel, 1992). Many patients with damage to the frontal lobe display unique characteristics on neuropsychological and behavioral tests including perseverations (i.e. th e inability to stop a behavior/response), intrusions, motor impersistence (i.e. an inability to sustain a motor gesture or action over a period of time), and poor self regulation (Alvarez & Emory, 2006; Helmstaedter, 2001). In addition, depending on the s ite of the frontal lobe lesion, patients can also display difficulties with expressive speech, motor weakness or incoordination, apathy, or problems with abstract thinking.

PAGE 14

14 While many of these characteristics are unique to persons with damage to the fron tal lobes, others are found commonly in other patient groups. This is the certainly the case when studying patients with frontal and temporal lobe epilepsy. Typically, patients with TLE localized to the left hemisphere (LTLE) display robust reliable neuropsychological impairments that are generally replicated across studies. These include deficits on tests on object and person naming, verbal comprehension, and learning and memory for both simple and complex forms of verbal material (Hermann, Seidenberg, Sch oenfeld, & Davies, 1997; Hermann et al., 1999). However, LTLE patients commonly exhibit difficulty not only with verbal learning and memory, but also with response inhibition, impulsivity, set loss, and difficulties with mental fle xibility and abstract thi nking These overlapping patterns of impairments on neuropsychological tests are likely due to a variety of factors, including the fact that patients with TLE often have propagation of abnormal electrical activity from the temporal to frontal regions, the reby causing potential impairment (Hermann & Seidenberg, 1995) Further, patients with TLE can exhibit reductions in white matter volume in frontal cortex, in addition to a reduction in overall cerebral volume and gray matter changes (Hermann et al., 2003 ; McMillan et al., 2004; Oyegbile et al., 2006). In addition, patients with longstanding seizure disorders may exhibit depressed cognitive profiles on multiple cognitive domains due to the cumulative effect of uncontrolled seizures (Jokeit & Ebner, 2002). Further, this lack of differentiation between frontal and temporal lobe epilepsy patients may also reflect the relatively poor specificity of our assessments for localized brain dysfunction. Given the relative overlap of these cognitive profiles, identifi cation of seizure localization is at best a complicated and ambigu ous task for neuropsychologists. Measurement of Neuropsychological Functioning In FLE and TLE As previously mentioned, patients with frontal and temporal lobe epilepsy may present with over lapping neuropsychological profiles. This may result, in part, from overlapping neural

PAGE 15

15 pathology, and in part because of relatively poor ability of clinical tests to discriminate frontal from temporal lobe epilepsy. A pre -surgical neuropsychological test battery for epilepsy patients includes tests of overall cognitive functioning, expressive and receptive language, verbal and nonverbal memory, processing speed, visuoconstructional ability, attention, executive function, and mood. Tests routinely used t o assess for executive or frontal lobe dysfunction usually include measures such as the Wisconsin Card Sorting Test (WCST) Stroop Color -Word Test (Stroop), Trail Making Test (TMT ), and measures of verbal fluency, including phonemic and semantic fluency tests ( Lezak et al., 2004). A dditional measures such as Lurias Motor Sequencing Tests (Luria, 1966) may also be administered in some clinics (Stuss & Levine, 2002) Combined results on these measures is then used to determine whether or not a patient di splays significant executive dysfunction and, in combination with other test results, whether or not th e patient displays a localizing pattern of deficits. The difficulty in clinical decision -making lies in the fact that these tests are sensitive to the pr esence of brain disease or damage, but may not be sufficiently specific to damage in a particular region of the brain. A body of evidence supports the notion that impairment on measures of executive function are not specific to damage of the frontal lobe, and do not discriminate frontal patients from those with temporal or other types of damage. In epilepsy populations, Hermann, Wyler, and Richey (1988) initially documented significant errors in planning and problem -solving on the Wisconsin Card Sorting Test (WCST) in a small group of left and right TLE patients compared to generalized seizure patients and controls. The finding of poor performance on the WCST in TLE, as measured largely by the number of perseverative errors, has been confirmed by others ( Martin et al., 2000; Trennery & Jack, 1994; Corcoran & Upton, 1993). A recent meta analytic review (Emory & Alvarez, 2006) found that out of twenty -five lesion studies, twelve

PAGE 16

16 found that adults with frontal lobe lesions performed more poorly on the WCST t han healthy controls, and 10/16 studies suggested that frontal lobe lesioned persons performed worse than those with extrafrontal lesions. However, they also reported that two studies found no differences between frontal lobe patients when compared to norm ative data, and nine studies found no significant differences between groups with frontal cortical lesions and those with lesions elsewhere in the brain (i.e. basal ganglia, diffuse lesions) Additional studies using the Stroop paradigm have found that n on-frontal epilepsy patients perform poorly on this measure. One recent study by McDonald and colleagues (McDonald et al., 2005) found that patients with left lateralized TLE were significantly impaired on measures of switching and inhibition on the Color -Word Interference Test (McDonald et al., 2005). Despite the frequency of use by neuropsychologists, few lesion studies have employed the Stroop paradigm to examine its specificity to frontal lobe lesions. Of those that have, only two studies found that per sons with lesions to the frontal lobes performed worse than controls (Stuss et al., 2001; Vendrell et al., 1995), and two studies reported those with frontal lesions perform worse than nonfrontal controls on the interference trial (Perret, 1974; Stuss et a l., 2001). On the other hand, studies have found the opposite pattern; frontal lobe and temporal lobe lesioned patients performed equally poorly on this measure (Blenner, 1993). Demakis (2004) conducted a meta analysis of studies which have employed the T MT and Category Test to determine the relative utility of the se instruments in detecting frontal lobe damage or dysfunction. Based on the studies included, 321 participants were included in the overall meta analysis, although sample size varied slightly by test examined. Surprisingly, the results of the study indicated that frontal patients performed significantly worse than non -frontal patients on Trails A (thought to assess mainly psychomotor speed), but did not perform worse on

PAGE 17

17 the Category Test or Trail s B. Exner and colleagues (Exner et al., 2002) found that patients with frontal and temporal lobe epilepsy had inferior Trails A and B performance compared to controls, but did not significantly differ from each other on these measures. In sum, these resu lts and others indicate that tests commonly thought to assess the integrity of the frontal lobe are in many cases, sensitive to frontal dysfunction. However, many studies have failed to find such effects, and other studies have found that, while some of th e tests may be sensitive to frontal lobe damage, they are nonspecifically impaired in damage to the temporal lobe and elsewhere in the b rain. These results raise questions about the utility of many so -called frontal -executive tests and at best, indicat e that such tests may be sufficiently sensitive but insufficiently specific These findings suggest the need to further develop neuropsychological methods with greater specificity for frontal lobe disturbance in clinical populations. Verbal Fluency Verba l fluency measures are among st the most common measures administered in traditional neuropsychological assessment (Stuss & Levine, 2002), as more than 50% of neuropsychologists report using these measures in standard clinical practice (Butler et al., 1991) Fluency measures are quickly administered, easily scored, and readily available. Further, adequate norms exist for these measures making their use in clinical practice even more prevalent. These measures require timerestricted generation of multiple re sponse alternatives under constrained search conditions and involves associate exploration and retrieval of words (Henry & Crawford, 2004). Although a variety of fluency measures exist (i.e. written fluency, figural fluency) the most common varieties of f luency tests are oral in nature, and assess word generation to either a phonemic or semantic cue. Phonemic fluency requires generation of words

PAGE 18

18 that begin with a particular letter (i.e. F, A, S, or C, F, L), whereas semantic fluency assesses word productio n to a given category (i.e. animals, items in a supermarket, fruits and vegetables). Semantic F luency Presently, there is disagreement in the literature as to the extent to which semantic fluency is sensitive to the integrity of the frontal lobes. Some evidence suggests that phonemic and semantic fluency may impose differing demands on frontal -executive processes; searching for semantic items within a larger superordinate category places demands on well -established search mechanisms that are congruent wi th organizational structures in our environment (e.g. generating items that can all be found in a supermarket, as opposed to generating items by letter, which contain no inherent semantic relationships). Perret (1974) has argued that because the search cri teria for semantic fluency are consistent with the natural organization of the human lexicon, the demands of this task rely less on the executive processes, and more on the integrity and organization of semantic memory stores. Others argue, however, that p atients with executive dysfunction are unable to perform effective and strategic searches through memory, irrespective of whether the search is semantically or phonemically driven (Baldo et al., 2006; Baldo & Shimamura, 1998; Troyer et al., 1998), and as s uch, would be equally impaired on semantic and phonemic fluency tests. Empirical data provides some clarity to the theoretical debate, although the body of literature is not entirely consistent. For instance, Drane et al. (2006) found that patients with frontal lobe epilepsy were more impaired than a group with temporal lobe epilepsy on measures of semantic fluency. A study comparing patients with focal anterior and posterior lesions found that both types of lesions produced impairments on semantic, or c ategory, fluency (Stuss et al., 1998). Additional studies employing a variety of populations have found similar impairments in semantic fluency in frontal -lobe patients (Baldo & Shimamura, 1998; Costello & Warrington,

PAGE 19

19 1989; Owen et al., 1990; Randolph et a l., 1993). Conversely, studies report spared sem a ntic fluency in patients with frontal lesions (Corcoran & Upton, 1993; Joanet & Goulet, 1986; Jurado et al., 2000, Vilkki & Holst, 1994). In addition, secondary interference tasks though to disrupt frontal lobe functioning have not been successful in disrupting category fluency performance (Mack, 1994; Martin et al., 1994 ). Functional imaging studies have attempted to delineate brain regions associated with successful category fluency performance. Mummery (Mummery et al., 1996) reported significant left temporal lobe activation in the inferior and anteromedial regions while patients performed category fluency tasks, but did not find significant frontal lobe activation for this task. Using voxel -based lesion mapping, Baldo and colleagues (2006) found that category fluency performance was associated with lesions in the left temporal lobe, post -central gyrus, parietal cortex, and putamen. When examining areas specific to category fluency the most important reg ions of interest were in the temporal (Brodmanns Areas (BA) 22, 37, 38, 41, and 43) and parietal cortices (BA 7, 39). No significant regions in the frontal lobe were noted (Baldo et al., 2006). These findings have been replicated by other functional imag ing studies (Gourovitch et al., 2001), but additional areas of activation have been found for category fluency, namely the left hippocampus and medial frontal cortex. It is commonly believed that semantic fluency performance relies heavi ly on the integr ity of intact semantic memory networks or the modules of long -term memory that contain knowledge about objects, concepts, facts, as well as the meanings of words largely localized to the temporal structures of the language dominant hemisphere (Butters et al., 1987; Monsch et al., 1992). Consistent with this perspective, lesion studies examining patients with temporal lobe involvement are generally impaired on tests of semantic fluency, compared to other patient

PAGE 20

20 groups and controls. Compared to healthy controls, Troster and colleagues demonstrated that patients with left TLE generated fewer words on a semantic fluency test (supermarket fluency) (Troster et al., 1995). This pattern of worse semantic fluency performance in LTLE compared to controls has been r eplicated by others (Gleissner & Elger, 2001; NKaoua, 2001; Martin, Loring, Meador, & Lee, 1990). A pattern of impaired semantic fluency performance has also been reported in other clinical samples with temporal lobe involvement, including Alzheimers Dis ease (Randolph et al., 1993; Diaz et al., 2004). A recent meta analysis of 995 patients (Henry & Crawford, 2002) with a wide range of lesion etiologies found that temporal patients were more impaired on semantic than phonemic fluency, and that those with l eft lateralized temporal lesions were more impaired than those with right temporal lesions. Results on the aforementioned neuroimaging, dual -performance, and voxel -based lesion mapping studies also confirm the crucial role the left temporal lobe plays in r etrieval of words from superordinate semantic categories. These findings indicate adequate performance on semantic fluency tasks is multi determined. It is clear that semantic fluency relies heavily upon access to the semantic memory stores of the tempo ral lobe, and that damage or disease processes involving this region impair s successful performance on semantic fluency. The role that the frontal lobes play in the controlled search process necessary to complete the task appears to be less critical, altho ugh damage to the frontal cortex can also impair semantic fluency to a lesser degree. Phonemic F luency Phonemic fluency measures commonly consist of three trials, which require generation of words beginning with a particular letter (Lezak, 2004). While phonemic fluency performance is obviously dependent on the integrity of language systems as well, this measure has traditionally been conceptualized as a measure of executive function because of the unusual demand of word

PAGE 21

21 generation based on orthographic c riteria. Further, the task requires the creation of nonhabitual strategies of word retrieval based on lexical representations, and the suppression of responses based on their meaning (Perret, 1974). Effective performance on this measure also requires effic ient organization of verbal recall, retrieval, and output, in addition to self -monitoring, effortful self-initiation, an inhibition of previously given responses (Henry & Crawford, 2004; Ruff et al., 1997). Research supports the assertion that there is a relationship between the integrity of the frontal lobes and performance on phonemic fluency. The finding of decreased phonemic fluency in frontal lobe patients has been reported in patients with traumatic brain injury (Jurado, et al., 2000), left frontal and bi -frontal epilepsy (Troyer et al., 1998), dementias involving the frontal lobes (Rosser & Hodges, 1994), and a variety of patient groups of mixed frontal lobe pathology (Stuss et al., 2000; Janowsky et al., 1989). In their meta analysis, Emory and Alv arez (2006) found that the majority of studies of frontal lobe lesion patients reported significantly poorer phonemic fluency scores compared to controls, although a smaller percentage found this same difference compared to non -frontal lobe lesions patient s. Another meta analysis reported large effect sizes ( r =.52) for deficits of their frontal lobe group compared to their non-frontal group, and deficits were largest with left frontal lesions, although patients with left focal non -frontal lesions also show ed significant impairment on phonemic fluency tests (Henry & Crawford, 2006). While the sensitivity of this measure has been demonstrated in lesion studies, its specificity has not yet been established because a number of studies demonstrate no significant differences between frontal patients and those with either non-frontal cortical or diffuse lesions (Stuss et al., 1998; Miller, 1984; Pendelton et al., 1982; Perret, 1974). Further, some authors suggest that while phonemic fluency does indeed tap an exec utive factor, the contribution of

PAGE 22

22 verbal abilities to overall performance are equally important. Indeed, Ramier and Hcaen (1970) argued that successful performance on phonemic fluency is determined by an executive factor located within the frontal lo bes and a verbal factor mediated more generally by the left hemisphere function (presumably the language dominant hemisphere). Neuroimaging studies of healthy controls confirm the critical role of the frontal lobes in phonemic fluency performance, despite significant variability in task procedure and imaging parameters. Studies have found specific areas of increased activation in the left inferior frontal gyrus (IFG), anterior cingulate (AC), and left dorsolateral prefrontal cortex (DLPFC) (Paulesu et a l., 1997; Frith et al, 1995; Frith et al., 1991). Other studies report activations in these areas while also finding significant increases in blood/glucose to more widespread areas of the frontal lobes (Parks et al., 1988). In sum, evidence from lesion and neuroimaging studies suggests that phonemic fluency relies upon the integrity of the frontal lobes, much more so than semantic fluency. Semantic fluency on the other hand, appears to be more sensitive to the integrity of the temporal lobes and places a larger demand on semantic memory stores. Nonetheless, contradictory reports in the literature draw the specificity of these measures into question, as it is clear that both frontal and temporal lobe patients can exhibit impairment on either, or both tasks At this point in time, the scientific literature does not provide definitive support for the notion that frontal lesions necessarily produce disproportionate impairment on phonemic fluency, and temporal lesions disproportionate ly affect semantic fluency performance It may be that further refinements in fluency tasks, or in ways in which fluency performance is measured, might improve the ability to provide such a double dissociation. Providing such refinements is one purpose of the current research.

PAGE 23

23 Nov el Techniques for Dissociating Frontal and Temporal Lobe Impairment Many previous studies have attempted to differentiate frontal from nonfrontal lesions on the basis of performance on experimental tasks designed to isolate particular aspects of frontal e xecutive dysfunction. These include assessments of set shifting (McDonald et al., 2005a), figural fluency (McDonald et al., 2005b), directed forgetting (McDonald et al., 2006), priming (Alexander et al., 2005; Stuss et al., 1999) self ordered pointing (La mar & Resnick, 2004; Petrides & Milner, 1982), source memory (Thaiss & Petrides, 2003), and structured semantic cueing paradigms (Drane et al., 2006; Randolph et al., 1993). This general tradition has also led to refinements of traditional oral fluency par adigms. One such paradigm is based on the literature that posits distinct neural regions for the processing of words that denote concrete entities, such as objects, and words that denote action or motion, verbs. Action/Verb Retrieval Although a fair amo unt is known about the neural representation for words denoting concrete entities (objects) less is known about the neural basis of action word retrieval. However, a growing body of literature across many fields, including linguistics, cognitive and expe rimental psychology, neurology, and clinical psychology lends support to the idea that distinct neural regions are involved in more highly specialized in processing information that relates to action or movement. Why should actions and objects have differe nt neural bases? One theory posits that knowledge about objects and actions is stored in association cortices adjacent to the primary cortical regions that process these classes of stimuli (Damasio & Tranel, 1993). According to this theory, object knowledge is stored in cortical regions adjacent to the occipito temporal visual stream, while action knowledge is stored adjacent to motor structures in the frontal lobe

PAGE 24

24 including the prefrontal cortex, premotor cortex, and supplementary motor area (Lu et al., 2 002) that process skilled movement and action. As such, it is suspected that the frontal lobes likely serve as a storehouse for knowledge related to movement/action, and to the extent that the major semantic features of an object include implications for m otion, a substantial portion of its neural network will reside within the frontal lobes. In contrast, object knowledge, based as it is on structural representations of object form, is more dependent on the occipitotemporal visual stream, which normally pr ocess object qualities. Another theory posits that words are segregated based on how they were learned, thus emphasizing the distinction between objects, which are highly visual, and actions that have salient functional components (Warrington & Shallice, 1984). Alternatively, because actions are captured by grammatically rich verbs, it is possible that difficulty recognizing and producing action words is related primarily deficits in grammatical processing (McCarthy & Warrington, 1985). Still another vie w holds that the deficit is largely executive in nature, and relates to the difficulty of mentally coordinating and manipulating the large amount of information related to action -words (White Devine et al., 1996). Although the outcome of this debate rema ins unclear at this point, it may be possible that several of these theories will eventually contribute to our understanding of the mechanism underlying category -specific deficits. Further, regardless of which theory is correct, all posit separate neural s ubstrates for objects and actions. T he literature is beginning to draw a clear picture that category -specific deficits for action words (verbs) do exist. The strongest support for this claim is from the lesion literature, in which a variety of studies ha ve demonstrated this category specific deficit. This finding was first noted in agrammatic aphasics who demonstrated notable deficits in verb production, whereas anomic aphasics had greater impairment in the retrieval of nouns (Miceli et al., 1984). Damasi o & Tranel

PAGE 25

25 (1993) present three compelling case studies; two of their patients had damage to the anterior and middle temporal cortices and the third in the left premotor cortex. Results revealed a double dissociation in which their first two patients had d ifficulty naming common nouns (pictures of objects) whereas the third patient was unable to name actions depicted in line drawings. A similar category -specific noun/verb effect has been shown in other groups of patients with damage to various regions with in the frontal cortex (Caramazza & Hillis, 1991; Daniele et al., 1994; Hillis & Caramazza, 1995; Miceli et al., 1984; Rapp & Caramazza, 1998). Exploiting the known neuropathology of various dementia types, researchers have also demonstrated action naming i mpairments in groups with frontal lobe pathology, and the lack of impairment in patients with an absence of this pathology. Although both Alzheimers (AD) and fronto-temporal dementia (FTD) patients displayed impairments in object and action naming compare d to controls, the discrepancy between object and action naming performance was significantly larger for FTD than AD patients regardless of dementia severity (Cappa et al., 1998). The same group later found that impaired action naming was not only present in FTD, but also in other patient groups with frontal -subcortical disease, including those with supranuclear palsy and corticobasal degeneration (Cotelli et al., 2006). Another study comparing action and object fluency in AD and FTD confirmed Cappas resul ts, but further elucidated the nature of the action naming disorder In FTD the naming disorder was found mostly to be due to a dysexecutive deficit whereas in AD, it was due largely to linguistic difficulties (Silveri et al., 2003). Although the possibil ity of a selective verb deficit has not been explored in frontal lobe epilepsy, one study did find that verb naming was spared in patients who had undergone LATL; a finding that is consistent with the view that action naming is not localized to the tempora l lobes (Glosser & Donofrio, 2001).

PAGE 26

26 Functional imaging studies in both patients and healthy volunteers further confirm the role of the frontal lobes in the retrieval of action words. In fact, one of the earliest tasks used in PET and fMRI studies was a ta sk in which participants were shown a picture of an object (ball) and asked to generate a verb (kick) for the object. These early imaging studies found that the left inferior frontal gyrus (IFG) was activated during these tasks (Petersen et al., 1989; Petersen, Fox, Snyder & Raichle, 1990; Raichle et al., 1994). More recent studies have confirmed the role of the frontal cortices in naming and generating action words. ThomsponSchill found that patients with damage to the left IFG not only had more diffi culty generating semantically appropriate verbs, but also made more errors on their task than did patient or elderly control groups (Thompson -Schill, 1998). Other fMRI findings using varying task demands found similar activations in the left inferior prefr ontal cortex (Perani et al., 1999; Shapiro, Moo & Caramazza, 2006; Tyler et al, 2004). One recent PET study found that naming actions was correlated with increased glucose utilization in the left frontal operculum, left posterior middle frontal gyrus, and left and right parietal lobule (Damasio et al., 2001). Until recently, the assessment of verb retrieval abilities has been limited to action naming paradigms, similar to an action analog of traditional naming tests, such as the Boston Naming Test (BNT). While these paradigms are useful in assessing verb naming impairments, they all require identification of a verb associated with a graphically depicted image (Obler & Albert, 1979) as opposed to free generation of action-related words. Over the past sever al years, however, a small body of literature using action fluency paradigms has emerged. Action fluency paradigms assess the spontaneous production of verbs, with the instructions tell me as many different things as you can think of that people do (Piat t, Fields, Paolo, & Troster, 1999).

PAGE 27

27 Preliminary evidence from action fluency studies suggests that performance on this task i s indeed sensitive to frontal lobe dysfunction. Patients with known frontal -subcortical damage were significantly impaired on act ion fluency when compared to healthy controls. P atients with frontal -subcortical pathology secondary to HIV 1 infection performed similarly to healthy controls on measures of semantic fluency, but were significantly impaired on the action fluency measure ( Woods et al., 2005). When Parkinsons patients with (PDD) and without dementia (PDND) and healthy controls were compared on action, phonemic, and semantic fluency tasks, PDD patients performed worse on all three measures. However, performance on the action fluency task was differentially more difficult for the PDD group than semantic or phonemic fluency, relative to the control and non-dementia groups (Piatt et al., 1999b). The authors conclude that the measure was both sensitive and specific to frontal -su bcortical disease. Three studies have examined the construct validity of the action fluency test as a measure of frontal lobe functioning and/or executive function. Piatt Fields, Paolo & Troster (1999) demonstrated the convergent validity in a sample of healthy older adults. They found that action fluency performance was significantly related to several measures of executive function (i.e. TMT B, WCST) but not with common measures of temporal lobe functioning (i.e. BNT, Logical Memory from the Wechsler Memory Scale). While action fluency shared common variance with other measures of executive function, the test also seemed to measure a component of executive functioning not tapped by more traditional tasks. Woods et al ( 2005) found similar relationships wit h putative measures of executive function in healthy young volunteers, but found no relationship with measures traditionally associated with the posterior cortex. In sum, there is support for dissociation between action and object naming, with the naming o f actions being dependent on anterior brain structures, namely the frontal lobes. Further, preliminary findings

PAGE 28

28 indicate that action fluency is a novel paradigm that may not only be sensitive, but also specific to frontal lobe functioning. To our knowledge this paradigm has been applied only to patients with HIV and Parkinsons Disease and normal older and younger controls, and has not been used in patients with focal epilepsy. Proper Name Retrieval The difficulty in isolating patients with frontal lobe damage lies not only in identifying tasks sensitive to frontal lobe dysfunction, but also in designing equiv alent tasks sensitive only to temporal lobe damage in order to doubly dissociate performance on neuropsychological test s. While the mesial temporal lobe is critically important in episodic memory, the lateral (cortical) aspects of the temporal lobes are likely critical in the storage and maintenance of semantic memory, or knowledge of objects, facts, and names P atients with damage to the anterior a nd lateral portions of the temporal lobes often have difficulty on tasks tapping semantic memory stores. This is particularly true in the case of proper names. The specific difficulty in producing proper names has generally been attributed to their semant ic uniqueness, or the fact that these names refer to unique entities (Semenza & Zettin, 1989; Grabowski et al., 2001). While common names refer to concepts, or a set of attributes that are shared by multiple entities within the same concept, proper names d o not inherently contain attributes in and of themselves and are merely expressions by which we refer to an individual person or item. Because of this, it is thought that widespread neural networks support the representation of common nouns, while proper nouns are thought to hold rather fragile associations with their unique reference (Martins & Farrajota, 2007). In addition, difficulty in retrieval of common nouns is often abated by the fact that they can often be substituted with synonyms, whereas this is not usually possible with proper names (Bredart, 1993).

PAGE 29

29 Regarding the neural representation of proper names, several cases studies have shown that lesions to the left anterior temporal and temporal polar regions selectively disrupt the retrieval and production of proper names (Damasio et al., 1996; Harris & Kay, 1995; McKenna & Warrington, 1980). Two lesion studies demonstrated this effect across naming paradigms (naming to pictures, naming to description), and name generation tasks (actors, sports f igures). The dissociation of common and proper name impairments was also demonstrated in a stroke patient (ACB) who suffered an ischemic lesion that involved that temporal neocortex and temporal pole (Martins & Farrajota, 2007) Subsequently, he was unabl e to recall proper names (particularly those that referred to well known figures, such as politicians ) while his ability to produce common names of objects was relatively spared. Similar deficits in proper naming abilities were documented in a patient who underwent left ATL surgery for refractory epilepsy (Fukatsu, et al., 1999). This patient was able to accurately perceive pictures of faces, but was able to name only 25% of his acquaintances from photos, and even fewer from verbal description (24%). He nam ed only 4 out of 25 famous faces correctly. He was able to name 90/100 pictures of common items, however (animals, furniture, tools, insects). Further, his fluency for common names was double his fluency for proper names. Similar, though less dramatic find ings of proper name impairment have been reported in patients who underwent LATL for epilepsy relief (Barr, Goldberg, Wasserstein & Novelly, 1990; Tsukiura et al., 2002; Seidenberg et al., 2002; Glosser, Salvucci, Chiaravalloti, 2003). In these studies, ma ny of the LATL patients had impaired naming of famous faces compared to controls, but were able to recognize or subsequently provide semantic information about them, apart from their actual name. Because the majority of these studies involve the naming of persons in response to visual representation, it raises the question of whether this impairment is related to the specificity of the

PAGE 30

30 task (i.e. retrieving unique names) or to preferential processing of facial stimuli by the anterior temporal lobe. Severa l neuroimaging studies have tried to address this question. Regardless of presentation format (printed names versus photo) or presentation modality (visual versus auditory), similar areas of activation are noted (Tempini et al., 1998; Tranel, Grabowski, L yon & Damasio, 2005). In addition, this question has been addressed directly by comparing retrieval of names of other unique entities (landmarks and buildings) to retrieval of proper names of persons. The results of these studies (Milders, 2000; Tranel, 2006; Grabowski et al., 2001) confirm the hypothesis that portions of the left temporal lobe (anterior temporal lobe, temporal pole) are in fact specialized for the retrieval of unique entities as a whole, not only entities that contain human features. Apar t from a limited number of lesion studies that have used the design experimentally, the assessment of unique (or proper) naming abilities has been limited to paradigms that require identification of a name associated with pictures or photographs. Although there is sufficient evidence to support the idea that the retrieval of proper names depends on the integrity of anterior portions of the temporal lobe, and taps a unique aspect of semantic processes, these hypotheses have not been directly tested in this f ormat in clinical populations, and the ability to generate proper names without corresponding visual stimuli has not been examined. Qualitative Analysis Tests of verbal fluency, regardless of type, are generally scored by tallying the total number of words generated, minus errors or repetitions. While this score is an accurate measurement of fluency output, it provides minimal information about how the task is completed, and as previously discussed, may be limited in accurately characterizing performances by different groups of patients because the same score can be obtained in qualitatively different ways. Because fluency score is likely multi -determined, and affected by

PAGE 31

31 difficulty initiating or maintaining performance, faulty search and retrieval strateg ies, degraded semantic memory stores, failure to maintain set, and self -monitoring failures, a more thorough analysis of the cognitive processes involved in task performance is warranted. Aspects of fluency performance, including perseverations and intrus ions, have been assessed in a variety of clinical and healthy populations (Reverberi, Laiacona, & Capitani, 2006; Warrington, 2000; Martin & Fedio, 1983; Troster et al., 1989) and recently, Troyer, Moscovitch, & Wincour (1997) developed a methodology for e xamining organizational retrieval processes involved in word generation. They suggest that optimal fluency performance is composed of the production of semantic (i.e. apples, bananas, grapes) or phonemically -related (i.e. far, fat, fast) clusters of word s, and when one cluster is exhausted, a switch is made to another cluster. As such, they envision two important aspects of fluency performance; clustering which is the production appropriate words within the subcategories, and switching the ability to sh ift between said subcategories. Clustering is thought to rely heavily upon organized access of semantic memory stores (more strongly localized in the temporal lobe), while switching is thought to rely more heavily upon cognitive flexibility, ability to shi ft set, disengagement from previous responses, and strategic search (more strongly localized in the frontal lobe; Troyer & Moscovitch, 2006). Some evidence suggests that while clustering relies on relatively automatic cognitive process switching is though t to involve effortful processing (Rende, Ramsberger, & Miyake, 2002). Data from healthy young and old volunteers has suggested that clustering and switching are indeed dissociable processes. Clustering and switching scores were both related to total flue ncy score on semantic fluency measures, but the switching score was more uniquely related to overall phonemic fluency, consistent with the view that both rely more heavily on frontal -

PAGE 32

32 executive processes (Troyer, Moscovitch & Wincour, 1997). The finding tha t divided attention tasks ( concurrent finger -tapping) disrupted switching but not clustering also supports this assertion. Given that these processes have been identified as dissociable in healthy volunteers and have shown differential relationships with overall fluency scores, others examined whether they would be differentially affected by neurological disorders. As predicted, patients with temporal lobe epilepsy showed decreased clustering on semantic fluency tests, while patients with frontal lobe epil epsy exhibited decreased switches on both phonemic and semantic fluency tests. Troyer et al. (1998a) determined that the best indices for discriminating these patients were phonemic switching and semantic -clustering scores. Early AD patients, who have know n temporal lobe pathology, showed reduced cluster size on both types of fluency (Troyer et al., 1998b). Patients with frontal and/or subcortical disease demonstrated the expected pattern of relatively intact semantic cluster size, but decreased switching ( Demakis et al., 2003; Ho et al., 2002; Troster et al., 1998) as did patients with psychiatric disease known to affect frontal -lobe functioning (Fossati et al., 2003; Robert et al., 1998). Quantitative fluency scores are certainly sensitive to frank pathol ogies such as those underlying Alzheimers Disease and aphasia, and can be also be sensitive to milder forms of pathology in some cases. However, impairment of this score can be due to heterogeneous causes, which greatly limits or precludes interpretation about the underlying cognitive processes responsible for the deficit. A qualitative analysis of fluency performance, however, helps to elucidate the mechanisms by which the task is completed and may shed light on the nature of fluency impairment. This qual itative analysis appears to be most helpful in identifying components of fluency performance due to frontal and temporal lobe impairment.

PAGE 33

33 Summary Elucidating a pattern of neuropsychological test performance characteristic of frontal and temporal lobe epi lepsy is complicated at best, but remains important for a number of reasons. First, identification of patterned impairments allows for better and more accurate identification of seizure onset, which is essential in treatment planning. Beyond the clinical and practical importance, a more thorough understanding of these cognitive impairments further extends our scientific knowledge about the neural substrate of these cognitive processes. Presently, these cognitive patterns are poorly defined and overlapping, partly due to the lack of specificity of many of our tests. In particular, many of our tests thought to identify frontal lobe dysfunction appear sensitive, but not specific. These include tests thought to assess mental flexibility, set -shifting, respon se inhibition, and working memory. Traditional measures of fluency are commonly used to identify patterns of performance of frontal and temporal lobe epilepsy patients, however, both patient groups may exhibit deficits on both types of tests, affording the m minimal discriminative validity. Despite the relative lack of specific standardized assessments, the scientific literature provides insight into the type of impairments that may exist with damage to either the frontal or temporal cortices. Specifically damage to the language dominant temporal lobe produces deficits in the retrieval or names of unique entities (people or places), whereas damage to the frontal cortex, particularly left lateralized damage, produces impairments in the retrieval of words de noting actions. In conjunction with the test content, a qualitative examination of the cognitive strategy involved may also useful predictive value. The current study seeks to incorporate this material -specific content into traditional test paradigms in order to further explore the unique cognitive deficits associated with localized

PAGE 34

34 neural dysfunction. By manipulating the retrieval demands involved and examining the cognitive strategies employed, we hope to more accurately discriminate between patient grou ps and to advance our understanding about the neural specificity of these brain regions. Specific aims of the study are listed below.

PAGE 35

35 CHAPTER 2 SPECIFIC AIMS & HYPO THESES OF CURRENT ST UDY Aim 1 Th e first aim of the study was to characterize performance o f patients with either frontal or left temporal lobe epilepsy, and matched healthy controls on a panel of verbal fluency tests, that includes both traditional measures of semantic and phonemic fluency, and experimental measures of action and proper name fl uency. We hypothesized that overall fluency score on a ct ion and proper noun fluency would doubly dissociate patients with frontal and temporal lobe epilepsy, with frontal lobe patients performing worse on action fluency and temporal lobe patients exhibitin g comparative deficits on tests of proper name fluency. We also suspected that patients with temporal lobe epilepsy would also evidence impaired semantic fluency (but not phonemic fluency), while patients with frontal lobe would demonstrate impairments on phonemic fluency (but not semantic fluency). All patient groups were predicted to generate fewer total words than controls due to the overall effect of their neurological disorder. Aim 2 The second aim of the study was to compare the clinical utility and predictive validity of experimental v ersu s traditional f luency measures in identifying seizure location and lateralization, and in discriminating between temporal and frontal groups. We predicted that the experimental fluency measures would have similar, a nd possibly more favorable operating characteristics (sensitivity, specificity, positive and negative predictive value) than traditional measures of fluency, and that our panel of fluency measures would be effective in accurately discriminating frontal and temporal lobe patients.

PAGE 36

36 Aim 3 The third aim was to establish the psychometric properties of the experimental fluency measures (convergent and discriminant validity) using traditional measures of frontal and temporal lobe dysfunction and to compare the pr edictive power of various fluency tests with more traditional neuropsychological measures. We hypothesized that performance on tests of c om mon and proper noun fluency would be related to measures of language and semantic/episodic memory (Wechsler Memory S cale -Logi cal Memory, Boston Naming Test), while action and phonemic fluency scores will exhibit moderate relationships with traditional measures of executive function (Wisconsin Card Sorting Test, Trail Making Test B). However, as these tests tap varied a spects of cognitive functioning whose neural instantiations exist within the frontal lobe, we thought that this fluency measure may comprise a new dimension of executive function not assessed by other measures. Aim 4 The four th aim was to determine wheth er a qualitative analysis of fluency performance (i.e., clustering and switching performance) would dissociate performance of patients with FLE, TLE, and healthy controls. Consistent with previous lesion studies, w e expected that patients with TLE would exhibit an average number of switches, but reduced cluster size, particularly on tests that rely more heavily on semantic memory (proper name fluency, common noun fluency). Conversely, we thought that patients with FLE would display the reverse pattern of spared semantic cluster size, but reduced number of switches, predominantly on tests of phonemic and action fluency.

PAGE 37

37 CHAPTER 3 SUBJECTS & METHODS Study Participants Participants in this study included patients with documented epilepsy localized either to the frontal or temporal lobes. A healthy control group with no current or past history of neurological disease was also recruited for participation in the study. Initial power analyses, computed from data provided in Troyer et al. (1997), indicated that wi th an evenly distributed sample of 30 (10 patients with left frontal lobe epilepsy, 10 patients with left temporal lobe epilepsy, and 10 healthy controls), our study would be powered adequately (Critical F = 3.55, Actua lity = .05) to detect overall group differences in fluency score, our primary aim of the study. However, additional analyses computed from data in Troyer et al. (1997) also suggested that as few as six participants were neede d per group (Critical F = Initial investigation of pre -surgical patient flow over the past several years suggested our patient recruitment goals were feasible in a pre -surgical population over a ten t o twelve month recruitment period. While the initial goal of the study was to recruit pre -surgical patients with either language -dominant temporal lobe epilepsy ( i.e., left hemisphere) or patients with left frontal epilepsy, this goal could not be achieved even with extended recruitment over a period of eighteen months. In eighteen months, we were successful at recruiting only five presurgical patients with left temporal lobe epilepsy who met all inclusion/exclusion criteria and four patients with frontal lobe epilepsy (one right, two left, one bifrontal). Because of the significant difficulty recruiting patients pre -surgically, post -surgical patient data was collected concurrently. Combining both pre -surgical and post -surgical patients, we were able to co llect data on fourteen patients with left temporal lobe epilepsy, seven patients

PAGE 38

38 with frontal lobe epilepsy, as well as eighteen patients with right temporal lobe epilepsy. By combining both pre and post -surgical data, we were able to meet our sample size requirements in our left temporal lobe group. However, even when combining pre and post -surgical data we were not able to obtain ten patients with left frontal epilepsy which was our initial goal Because of this, we combined patients with left frontal, right frontal, or bifrontal epilepsy to compose our frontal lobe group for a total of seven frontal lobe epilepsy patients (Table 3 1) Participants included in the final study were fourteen patients (pre and post -surgical) who had epilepsy localized by EEG to the left temporal lobe, seven patients (pre and post surgical) with EEG documented epilepsy of the frontal lobes (left, right, or bilateral) and twenty healthy controls. Lateralization and localization of seizure foci was determined by consensus diagnosis using data from Phase I EEG monitoring, MRI, patient history, and cognitive test results. At the present time, six of the nine pre -surgical patients have proceeded to surgery. Differences in demographic and clinical characteristics were assessed using one -way analysis of variance (ANOVA) and chi -square tests. Patients and control groups were well matched for age (F (2, 40) = .08, p >.05) education ( F (2, 40) = 3.1 p >.05), gender ( X2 (2) = .03 p >.05) handedne ss ( X2 (2) = 4.09, p >.05) race (X2 (2) = 3.51, p >.05) and WASI full scale IQ ( F (2,40) = 2.62, p > .05) (Table 3 2) On average, our patient and control groups were around forty years of age, had slightly more than a high school education, and were composed largely of Caucasians. While cont rols had slightly higher full -scale IQs, patients and controls did not differ significantly and o verall intellectual functioning for all three groups was in the average range. The groups were predominantly right handed (X2 (2) = 4.09, p >.05) and were com posed of slightly more women than men. Patient demographics are summarized in Tables 3 1 and 3 2.

PAGE 39

39 Patient groups did not differ significantly on language dominance (t (20) = .80, p >.05) or age of seizure onset ( t (20) = .28, p > .05). In general, temporal l obe and frontal lobe patients first developed seizures during their teenage years, though there was a wide range in the age of seizure onset amongst the patients (LTL M =16.9, SD =15.2; FL M =13.5, SD =11.7) (Table 3 2) Four of our TL patients and two of our FL patients had a history of psychiatric illness or diagnosis ( X2 (1) = .04, p >.05). With regard to risk factors for epilepsy, t hirty -five and twenty eight percent of TL and FL patients had a family history of epilepsy, respectively ( X2 (1) = .10 p >.05). Four TL patients had suffered a mild to -moderate traumatic brain injury (TBI) and three of the FL patient s experienced a mild to -moderate TBI ( X2 (1) = .2.14, p >.05) Slightly more TL patients had a history of child illness, though this was not statistica lly significant ( X2 (1) = 3.28, p =.07) In general, the patients in this study did not have a history of febrile seizures ( X2 (1) = 2.14, p >.05). Results of magnetic resonance imaging scans (MRI) revealed that ten of the TL patients had lesions consistent with mesial temporal sclerosis (MTS), and four of our FL patients had lesions or other neural anomalies ( X2 (1) = 5.57, p =.06). Epilepsy Patients Pre -surgical p atients All pre -surgical patients selected for inclusion in this study had medication refrac tory epilepsy, documented by a board-certified neurologist at the University of Florida Comprehensive Epilepsy Program (UFCEP), and had experienced uncontrolled seizures under at least two medication regimens for at least an 18 -month period. Participants w ere identified and recruited from the Neurology and Psychology Clinics at Shands Hospital and were screened to determine if they met inclusion criteria for the study, as approved by the Institutional Review Board (IRB) at the University of Florida. Inclusi on criteria for all patients were : 1) 18 years of age or older, 2) documented epilepsy of either the temporal or frontal lobes and 3) English as a

PAGE 40

40 first language. Exclusion criteria were : 1) history of severe developmental dis order, or mental retardation resulting in IQ < 69, 2 ) history of Axis I psychiatric disturbance that resulted in inpatient hospitalization, 3) history of substance abuse / dependence, using DSM IV criteria, 4) first language other than English, 5) lesional temporal -lobe epilepsy, or 6) presence of neurological disease in addition to epilepsy (e.g., radiation or chemotherapy for brain cancer in the past year; head trauma resulting in moderate to severe brain injury ). Patients were approached during either their Phase I ( inpatient ) vid eo EEG hospital stay or during their outpatient appointment in Neuropsychology, were presented with information about the study, and were asked if they would like to participate. Patients who agreed to participate gave informed consent and were tested in their hospital room as an inpatient in Shands hospital, or arranged testing on another day as an outpatient. In a few instances, testing could not be completed in one session and the patient was seen twice. R elevant demographic and medical information was also gathered from the participants medical record. Relevant d emographic information collected include d age, gender, educational and occupational attainment, reported handedness, and ethnicity. Medical and seizure related variables were also recorded, a nd included age of first seizure, current medications regimen seizure frequency and duration and Wada memory and language dominance. Over the course of eighteen months, a pproximately 150 pre -surgical patients were screened for the study via medical reco rd review. Of the patients screened, we recruited 37 patients for the study, and of those patients, nine met full inclusion criteria for the present study; five were determined to have clearly lateralized left temporal lobe seizures and four were determine d to have frontal lobe seizures (one right, two left, one bilateral). The remaining patients were determined to have non -epileptic seizures, seizures originating from another foci,

PAGE 41

41 had right or mixed language dominance, were unable to be classified based o n EEG recordings, or did not meet inclusion/exclusion criteria. Classification of seizure onset was determined by a board -certified neurologist who reviewed EEG recordings, clinical notes, and patient history. A summary of pre -surgical patient characterist ics is included in Table 3 1 and Table 3 2. Post -s urgical p atients Post -surgical epilepsy patients were patients who had undergone surgery between 2000 and 2007, and had undergone either an anterior temporal lobectomy or a frontal cortical resection Thes e patients were recruited from the Departments of Neurosurgery and Neuropsychology at the University of Florida. Patients were identified through clinical databases and were selected for recruitment based on surgery type, medical record review, and date of surgery. Patients were recruited throughout the entire state of Florida. Eligible patients were contacted via letter and provided with information about the purpose of the study, and its potential benefits and risks. They were provided with a self addres sed stamped postcard so they could indicate a desire to participate in the study, decline participation, or inquire further about study details. Patients who returned the postcard and were interested in participating or wanted more information were then co ntacted and given the opportunity to inquire further or schedule an appointment for cognitive testing. They were also screened briefly over the phone, after which the cognitive testing was arranged. Testing took place either at Shands Hospital, or if the patient was unable to travel to Gainesville at the patients home. We screened approximately 1,000 patients who had undergone epilepsyrelated neurosur gery (i.e., grids, resections, v agus nerve stimulator placement) between the years 2000 and 2007, and attempted to recruit approximately 125 by letter. The 125 persons contacted met

PAGE 42

42 criteria for study enrollment and had current mailing addresses. Of the 125 contacted, approximately 50 responded with an interest in the study. Of the 50 who expressed interes t, 30 were enrolled in the study. The remaining 20 patients either declined participation, were unable to be scheduled due to distance (> 5 hours away), or had travel, family, or work conflicts. Patients who participated in the study included nine patients with left temporal lobe resections, eighteen patients with right temporal lobe resections and three patients who had undergone right frontal cortical resections. A summary of post -surgical patient characteristics is presented in Tables 3 1 and 3 2. Hea lthy c ontrols We attempted to recruit family members of patients to serve as h ealthy volunteer (i.e. control) participants When patients were recruited into the study, family members who were present were also informed about the study and given the option to be screened and participate as a member of the healthy control group. Post -surgical patients were informed on the phone about the option of family members participating. These individuals were often tested the same day as their family member, in a quie t room in Shands hospital, although several were tested on alternative dates. The remainder of our healthy controls were recruited in Gainesville, F lorida through the use of fliers When interested persons called to inquire about the study, they were infor med of its purpose and procedures. They were also given a brief screening measure over the phone, consistent with IRB protocol, to determine if they were eligible to participate. Volunteers who were eligible and interested in participating engaged in a s in gle testing session in a quiet testing room on the ground floor of Shands Hospital. During this session, volunteers first gave informed consent in person, and display ed an understanding of the testing procedures.

PAGE 43

43 We attempted to match this group on age and education variables with our patient groups. Exclusion criteria for our controls were : 1) Age younger than 18, 2) history of severe developmental disability or mental retardation resulting in IQ < 69, 3) history of significant Axis I psychiatric disturban ce that resulted in inpatient hospitalization, 4) history of substance abuse or dependence, 5) neurological illness (e.g., epilepsy, cerebrovascular disease, brain tumor, or head trauma resulting moderate to severe head injury ), 6) current radiation or che motherapy treatment (within 1 year), or 7) English as a second language. We recruited twenty healthy controls to serve as volunteers in our study, eight of whom were family members of patients. A summary of control characteristics is included in Table 32 Measures Pre -surgical Patients Typically, pre -surgical epilepsy patients are routinely administered a standard neuropsychological test battery (SNB) as part of their Phase I evaluation through the UFCEP. Most patients in our study were administered the S NB, though some were still in the process of their pre -surgical evaluation, and had not yet undergone clinical neuropsychological testing. As part of the SNB evaluation, subjects were administered tests of verbal and non -verbal memory, naming, fluency, ex ecutive function, visuoconstructional and visuospatial ability, attention, mood, and intellectual functioning. A list of measures traditionally admi nistered is listed in Appendix A Because many measures of interest are routinely administered to patients, and because re testing patients on the same tests may have produce d biased results, data for selected tests from the SNB was used in the current study. Measures of particular interest from the SNB include d assessments of verbal memory (Logical Memory Stor ies from the Wechsler Memory Test -III [WMS III]), language (Boston Naming Test [BNT]), and measures of executive

PAGE 44

44 function (Wisconsin Card Sorting Test [WCST], Trail Making Test B [TMT B], and WAIS III Digit Span. As a detailed description of many of thes e tests has been provided elsewhere (Lezak et al., 2004; Spreen & Strauss, 2006) and are reviewed briefly in Appendix A they will not be described in further detail here. The study measures were administered during a separate experimental testing session These measures included a standard assessment of phonemic fluency, using three 60 -second trials of word generation beginning with the letters C, F, and L. These phonemic probes were chosen to be distinct from the phonemic fluency trial (FAS) given during the SNB. Analyses were computed using the letter C only so that overall score would be comparable to that of other fluency measures (i.e., a single trial) although analyses with overall phonemic fluency score (i.e., C+F+L) were computed as well T he semantic fluency category (i.e. Supermarket items) was also chosen for its non-overlap with category, Animals used in the SNB As a hierarchically organized supercategory, supermarket items also permit an analysis of how patients fluency performance reflects retrieval from multiple semantic subcategories within a trial. Total score for this was also the number of correct words generated within 60 seconds. Clustering and switching was also evaluated, as described below. Two experimental measures of fluency were administered. These measures are similar in administration time and scoring format to the previous measures, but differ in content. For the Actio n Fluency Test, participants were instructed to List as many different things that [you] can think of that people can do. The total score for this task is the number of correct target words, minus words denoting actions not performed by people (e.g., molt or photosynthesize ). Questionable answers (homonyms, i.e. bear -bare), or words with ambi guous grammatical roles (e.g., table) w ere queried by the examiner to determine if the word was an intrusion, and was

PAGE 45

45 scored accordingly. For the Famous Name Fluency Test, patients w ere be instructed to List as many different names of famous or well known people that you can think of, within a minute. Total score for this measure represented the number of correct names generated, minus names that c ould not be verified as famous persons, or intrusions or repetitions. As the format for these tests is simila r, the administration of these four trials w ere counterbalanced across participants to account for any practice effect related to familiarity with test administration procedures (i.e. ABCD, DCBA, BDAC, CADB). Qualitative analysis of the fluency protocols was completed in accordance with procedures described elsewhere (Troyer & Moscovitch, 2006). Briefly, clusters on phonemic fluency were defined as successively generated words that begin with the first two letters, differ only by a vowel sound, rhyme, or are homonyms. Semantic clusters were successively generated words that belong to a semantic subcategory, such as fruits, vegetables, meat, dried goods, and dairy, on supermarket fluency. Pilot data for our study revealed several naturally occurring cluste rs for action and name fluency. Clusters that emerged for action fluency were actions with the hands, feet/legs, facial gestures, grooming, household actions, work actions, emotional actions, actions involving language, and actions involving rest/relaxation Clusters on famous name fluency were actors/actresses, politicians, sports figures, TV personalities, and musicians/singers. Additionally, on this test clusters could be defined temporally (TV personalities from the 1980s) or by relationships (either personal or professional). As is customary, t he cluster size was counted beginning with the second word within each cluster. See Appendix B for examples. Switching scores were defined as transitions between clusters Although some controversy exists around this scoring procedure, the raw number of switches is used as the score, rather than a score corrected for total words produced. Because switching is in part, responsible for overall

PAGE 46

46 score, adjusting switches according to the total words generated would be akin to correcting a cause for its effect (Troyer & Moscovitch, 2006). To further emphasize this point, only raw switching score has been show to produce meaningful data (Troyer et al., 1998b). Errors and intrusions were also included in scoring cluste rs and switches, as they are thought to provide useful information about the underlying cognitive strategy, but are always excluded from total overall fluency score. All fluency measures were double -scored to ensure accuracy in calculation of total score, errors, clusters, and switches. Finally, apart from measures of cognitive functioning, all study participants w ere administered a brief questionnaire assessing demographic variables and the Edinburgh Handedness Inventory (Oldfield, 1971). Additional Mea sures In addition to the SNB and experimental measures listed above, two additional measures were given to study participants, The Action Naming Test (Obler & Albert, 1979), and a modified version of the Famous Faces Naming Test (Seidenberg et al., 2002). While these measures were not related directly to the primary aims of the study, they were administered to study participants for several reasons. First, the experimental fluency paradigms proposed for this study were based in part on the naming literatur e which provides evidence for the category -specific deficits for actions and proper nouns/names. This body of naming literature posits distinct neural substrates for action naming (i.e. the frontal cortices) and person naming (i.e. the anterior temporal cortices of the left hemisphere). While there is similar (but preliminary) support for a comparable substrate for action fluency, this paradigm has not been used extensively in the literature, and has not been used in epilepsy populations. Further, the name fluency paradigm has not been used widely either in clinical or research arenas, apart from its use in an isolated number of case studies.

PAGE 47

47 While we hypothesized these deficits w ould be apparent in a fluency paradigm, this extension has not yet been documen ted empirically. As such, the administration of naming analogues to our fluency paradigms allowed us to examine the pattern of deficits across naming and fluency tests in order to determine whether these unique category-specific deficits were isolated to n aming, or are in fact, present in a fluency format as well. Further, to our knowledge, the discrepancy between action and proper noun naming has not been explored in clinical epilepsy. The addition of these tests to our battery not only allow ed for compar isons between naming and fluency paradigms, but also provide d interesting information about various category -specific naming deficits which have been relatively unexplored in this population. Additionally, these naming tests were examined in conjunction wi th performance on a common object naming test, the BNT. A brief summary of these tests follows: Action Naming : The Action Naming Test (ANT; Obler & Albert, 1979) contains a series of 55 black and -white line drawings and was modeled after the B oston N aming T est (BNT). The items in this test, however, are a series of line drawings depicting actions. Sample actions include running, swimming, reading, curtsying, and exercising. Test items range in difficulty in ascending order. Study participants were shown on e picture at a time and we re instructed to tell what is happening in the picture, preferably with one word responses. Similarly to the BNT, phonemic cues were given if the participant could not name the action, although successful naming of these items a re scored as incorrect. Total score was the number of correct answers Famous Face Naming : The Famous Face Naming Test (Seidenberg et al., 2002) is a test that contains 100 black and -white photographs of famous or well known persons. A shortened version, containing 48 stimuli, was used for this study. The sub-set of stimuli was selected in

PAGE 48

48 order to shorten test administration time, and make it comparable in length to our other naming tests. Stimuli were excluded from the study if they contained features th at aided in identification (i.e. shirts with sports logos, war uniforms) or if the picture quality was deemed unsatisfactory. The stimuli are head -shots of famous persons sam pled across decade s since the 1960s, and contain images of athletes, presidents, actors, singers, politicians, and other newsworthy individuals. Sample faces include Kobe Bryant, Peter Jennings, Bob Barker, and Sting. Participants were initially asked if they recognized the person pictured and asked to provide any descriptive informa tion about them (i.e., a comedian; the boxer who bit someones ear off ; the guy who hosts the New Years Eve celebration in Times Square ). They were then instructed to provide the name of the person in the picture if able. No cues were given. Because recognition of test items varied across participants, the total score was corrected for the number of famous faces recognized ( i.e., correct names produced/recognized multiplied by 100; Drane et al., 2008). Post -s urgical Patients and Healthy Controls Healthy volunteers and post -surgical patients were administered identical measures. This included four fluency tests (phonemic, semantic, action, and proper name), and two naming tests (action and famous faces). Because healthy controls and post -surgical patients had not been administered the SNB, additional testing was completed in order to obtain scores on measures of language, verbal memory, and executive function. Accordingly, they were administered the WASI, WMS -III LM, WCST, TMT B, Digit Span from the WAIS III, and BNT. Pre -surgical, post -surgical and control participants were all compensated $10.00 per hour of time and given a $3.00 parking voucher. Average time spent in completion of this study was an hour for pre -surgical patients and twoanda half hours for post -surgical and control participants.

PAGE 49

49 Table 3 1 Patient seizure characteristics EEG Interpretation/Resection Site Surgery? Temporal Pre left temporal lobe and NES No Pre bilateral (left > right) temporal Yes Pre left temporal lobe Yes P re predominantly left temporal; 1 bilateral onset Yes Pre left temporal lobe Yes Post LATL 2001 Post LATL 2006 Post LATL 2004 Post LATL 2003 Post LATL 2002 Post LATL 2003 Post LATL 2006 Post LATL 2003 Post LATL 2005 Frontal Pre left fron tal lobe Yes Pre right frontal lobe Yes Pre bilateral frontal (left > right) No Pre left frontal lobe No Post Right Frontal Cortical Resection 2007 Post Right Frontal Cortical Resection 2006 Post Right Frontal Cortical Resection 2005

PAGE 50

50 Table 3 2. De mographic and clinical charac teristics LTL FL Controls N 14 7 20 Age (years) 41.1 (13.2) 39.4 (12.9) 40.9 (15.7) Education (years) 13.5 (2.6) 12.7 (2.5) 15.3 (2.7) Gender (M/F) 6/8 3/4 8/12 Handedness (R/L) 12/1 6/1 11/1 Race (Cau./AA/His.) 12/1 7/ 0/0 16/3/1 Full Scale IQ 96.1 (12.3) 94.0(14.6) 104.1 (9.8) Language Dom. (L/R) 14/0 5 /1 Age of Seizure Onset 16.9 (15.2) 13.5 (11.7) Note: Means are presented, standard deviations in parenthesis.

PAGE 51

51 CHAPTER 4 RESULTS The results of our analyses ar e presented below as they relate to the specific aims of the study: Aim 1 The first aim of our study was to characterize performance of patients with either frontal or temporal lobe epilepsy, and matched healthy controls on a panel of verbal fluency measur es including traditional measures of semantic and phonemic fluency, and experimental measures including action and proper name fluency. Because we were interested in determining whether a panel of fluency tests incorporating traditional and experimental measures would detect group differences in our patient groups, we first computed separate multivariate analysis of variance tests (MANOVAs) for the tests we hypothesized would be sensitive to TL pathology (semantic and name fluency tests) and the tests hy pothesized to be sensitive to FL pathology (action and phonemic fluency tests). Using Pillais trace, there was not an overall group difference for our patients in fluency performance on phonemic and action tests ( V =.02, F (2, 18) = .135, p There was, however, an overall significant group difference for the semantic and name fluency panel ( V =.28, F (2, 18) = 3.54, p Results of the omnibus MANOVAs are presented in Table 4 2. Univariate test results are presented below. When comp aring patients and controls, healthy controls outperformed patients on all measures of fluency including standard measures of semantic (supermarket) and phonemic (letter fluency) fluency, as well as experimental measures of fluency, including action and fa mous name tests. Overall univariate tests of analysis of variance ( ANOVAs ) revealed significant main group effects for action ( F (2, 38) =7.90 p =.001, ) and name fluency ( F (2, 38) = 17.02

PAGE 52

52 p =.000, ) tests, and there were trend s for significance for the semantic ( F (2, 38) = 2.94, p ) and phonemic fluency ( F (2, 38) = 2.73, p =.07, 35) measure s Planned post hoc contrasts revealed significant group differences between controls and both patient groups on action fluency (LTL < Control, p <.005; FL< Control, p < 05) and name fluency (LTL< Control, p <.0001; FL< Control, p <.01). When combining scores across phonemic fluency trials ( C p lus F and L trials ) healthy controls outperformed TL patients ( F (2, 38) =5.08, p =.011, ), but the difference in score was not significant for controls compared to FL patients. Contrary to our hypothesis, p erformance differences between frontal and temporal grou ps w ere not statistically significant on any of the four fluency measures However, there was a large effect size for group differences between TL and FL patients on name fluency ( d =.75), with FL patients generating more names than TL patients Means and standard deviations for patients and control groups are presented in Table and Figure 4 1. Repeated -measures ANOVA revealed that both FL ( F (3, 18) = 12.37, p <.001) and TL ( F (3, 39) = 30.72, p <.001) patients performed best on semantic fluency, followed by action fluency, phonemic fluency, and name fluency. FL patients performed significantly better on semantic fluency as compared to phonemic (p =.005) and name fluency (p =.009) TL patients performed significantly better on semantic fluency compared to action (p =.002) phonemic (p =.001) and name fluencies (p =.001) TL patients al so performed worse on name fluency compared to action fluency ( p =.01). There was a trend for significant differences between name and phonemic fluency ( p =.06) though this finding did not reach statistical significance (Figure 4 1)

PAGE 53

53 Aim 2 The second aim o f the study was to e xamine the clinical utility and predictive validity of experimental fluency measures in identifying seizure location and discrim inating between patient groups. Before attempting to dissociate patients groups, we sought to determine how well performance on our panel of fluency measures would dissociate patients from controls. Because our variables were not significantly collinear ( Supermarket: Tolerance=.771, VIF=1.29, Action: Tolerance=.46, VIF = 2.13, Name: Tolerance=.48, VIF =2.10, Phone mic: Tolerance=.51, VIF =1.95) we used forward -entry logistic regression with all four fluency measures included to predict patient status. This overall model was significant, and a good fit for the data ( X2 (4) =27.9, p<.001). This model correctly classifi ed seventeen healthy controls as controls, and classified eighteen patients as patients. The model incorrectly classified three controls as patients and three patients as controls, resulting in an overall classification accuracy of 85% Sensitivity and sp ecificity were both 85%. We then calculated multiple logistic regressions to determine how well differences in fluency performance predicted patient group membership. Again, we initially used forward -entry logistic regression with all four fluency measures included as predictors Entry of all four fluency variables into the model revealed a non -significant overall effect ( X2 (4 ) = 7.66, p =.10) and a poorly fitted model This model correctly classified twelve patients with TLE as TLE patients, but incorrect ly classified two TLE patients as having FLE. T his model also correctly classified three FLE patients as FLE patients, though incorrectly classified four FLE patients as TLE patients. Overall 86% of TLE patients were accurately classified, whereas 43% of FLE patients were accurately clas sified, resulting in a total correct classification accuracy of 71% (Table 4 3) Because of the poor fit of the model, we subsequently used backwards entry logistic

PAGE 54

54 regression to determine suitability of the variables for t he model. Results of this analysis provided two additional models. The first model included action, name, and semantic fluency, and showed a trend for significance ( X2 (3 ) = 7.66, p =.053). The final model fit the data well (X2 (2 ) = 7.66, p =.02) and inclu ded on ly semantic (p =.07) and name fluenc y (p =.053) removing the non -significant action (p =.91) and phonemic fluencies (p =.97) This final model correctly classified twelve TLE patients as having TLE, and incorrectly classified two as having FLE. This mod el also correctly classified five patients with FLE as having FLE, and incorrectly classified two as having TLE, resulting in an overall correct classification rate of 81%. This model correctly classified 86% of patients with TLE as having TLE, and 71% of patients with FLE as having FLE. Classification statistics for both models are presented in Tables 4 3 and 44. The predicted probabilities from this final model were subsequently used to predict a receiver operating characteristic (ROC) curve (Figure 4 2 ). Using the final predictors in our model ( semantic and name fluency tests only) the overall predicted area under the curve was .806. Aim 3 The third aim of the study was to e xamine the convergent and discriminant validity of the experimental fluency te sts with other tests sensitive to frontal and temporal lobe dysfunction and to examine the relationship amongst fluency measures. We first examined the relationship amongst fluency measures for patient groups only and then for patients and controls combi ned. When patients and controls were combined, there were positive, statistically significant correlations amongst all fluency measures. Semantic fluency was significantly correlated with name fluency ( r =.407, p =.008), action fluency ( r =.446, p =.003), and phonemic fluency ( r =.382, p =.01). Name fluency was also significantly positively correlated with action ( r =.662, p <.001) and phonemic fluency ( r =.628, p <.001). Performance on

PAGE 55

55 action fluency was also correlated with phonemic fluency performance ( r =.637, p <. 001). The correlation matrix for patients and controls is presented in Table 4 6. When we examined patients alone, there were no statistically significant correlations between semantic and phonemic fluency ( r =.076, p >.05), name and phonemic fluency ( r =.14 4, p >.05), semantic and action fluency ( r =.210, p >.05), or name and action fluency ( r =.285, p >.05) However, there were trends for significance for some correlations. As predicted, semantic and name fluencies were positively correlated with a moderate effect size ( r =.415, p =.06) There was also a positive correlation of moderate effect between action and phonemic fluencies ( r =.390, p =.08). These results are presented in Table 4 5. We also examined the relationship between performance on fluency tests and performance on traditional neuropsychological measures thought to b e sensitive to the presence of language dominant TL and FL dysfunction. These included the Boston Naming Test (BNT) and Logical Memo ry (LM) I and II (WMS III), and Digit Span (DS) (WAIS III), Wisconsin Card Sorting Test (WCST), and Trails B (TMT B), respectively. Measures of particular interest included the number of categories correctly completed and number of perseverative responses on the WCST, and total time to completion and total errors on the TMT -B. We examined these relationships separately for patients, and then for patients and controls combined When patients and controls were combined there were statistically significant positive correlations between semantic and name fluency and scores on the BNT, LM -I, and LM II, such that better performance on fluency tests was associated with better performance on the se measures ((semantic and BNT: r =.449, p <.01; semantic and LM I: r =.542, p <.001; semantic and LM II: r =.601, p <.001) and (name a nd BNT: r =.582, p <.001; name and LM I: r =.624, p <.001; name and LM II: r =.638, p <.001)). Semantic fluency performance was negatively correlated

PAGE 56

56 with TMT B time ( r = .382, p <.05) and TMT B errors (r = .455, p <.01) Finally, better name fluency was significant ly associated with better performance on DS ( r = .401, p <=01). Action and phonemic fluency were also positively correlated with performance on the BNT, LM -I, and LM II. A ction fluency performance was significantly positively correlated with DS total score ( r =.493, p <.01) and WCST Categories (r =.429, p =.01). Action fluency was significantly negatively correlated with TMT B time ( r = .485, p <.01) and WCST perseverations (r = .381, p <.05), such that persons who performed worse on this test had more perseverations on the WCST and took longer to complete TMT B. Phonemic fluency scores exhibited a similar pattern ; significant correlations were found between phonemic fluency and DS total score (r =.527, p =.001), TMT B time ( r = .430, p <.01), and WCST -perseverations ( r = .392, p <.05). When examining patients separately, semantic fluency remained positively correlated with LM II (r =.467, p =.05) and there was a moderate positive correlation and trend for significance with LM I (r =.416, p =.08). There was also a moderate neg ative correlation and trend for significance with TMT -B errors (r = .424, p =.07) Additionally, a ction fluency remained negatively correlated with TMT B time (r = .453, p <.05), and positively correlated with BNT (r =.543, p <.05) and LM II (r =.476, p <.05). The re was a trend relationship between action fluency and WCST -categories ( r =.448, p =.07). Scores on phonemic fluency were positively correlated with DS total score ( r =.494, p <.05), negatively correlated with TMT B time ( r = -.510, p <.05), and there was a trend for significance for WCST perseverations ( r = -.408, p =.12) and BNT (r =.415, p =.08). There were no significant correlations with name fluency. Correlation matrices are presented in Tables 4 7 and 4 8. Aim 4 The fourth aim of our study was to d etermine whet her a qualitative analysis of fluency performance, including the numbe r of clusters, cluster size, and switches, would dissociate

PAGE 57

57 performance of FL and TL patients Additionally, we sought to ascertain whether TL and FL patients display ed fewer clusters an d switches on fluency tests thought to most sensitive to temporal and frontal lobe dysfunction, respectively. We examined the number of clusters, switches, and mean cluster size for each individual fluency measure separately. On the semantic fluency task, controls, TL patients, and FL patients generated a similar number of clusters ( F (2, 36) = .574, p F (2, 36) = 3.14, p =.06 38), and did not differ significantly on the size of clusters generated ( F (2, 36) = .57, p fluency, with no sign ificant group differences on number of clusters ( F (2, 38) = .05, p =.94, 06), switches ( F (2, 38) = 2.87, p =. 07, 36), or cluster size ( F (2, 38) = .40, p =.67, =. 14). On action fluency clusters, significant group differences were found, ( F (2, 36) = 4.1, p planned contrasts revealed that controls generate d significantly more clusters than TL patients ( p =.05), but not FL patients. Controls also switched between clusters more frequently than patients ( F (2, 36) = 4.9 p =.0 1 4 5 ), with planned contrasts revealing controls switched more frequently than TL patients ( p =.01) There were no significant group differences in action fluency cluster size ( F (2, 36) = 2.25, p On name fluency, there were significant group d ifferences for total number of clusters ( F (2, 36) = 3.1, p =.05, F (2, 36) = 7.75, p <.01, .54 ), and a trend towards significance for mean cluster size ( F (2, 36) = 2.45, p Planned contrasts revealed that TL patients generated significantly fewer name clusters than controls ( p =.05) and that both TL ( p <.01) and FL ( p <.05) patients switched less frequently than controls. W hile not statistically significant, there was a large effect ( p =.16, d =.71) showing that TL patien ts also generated fewer clusters than FL patients on name fluency. Despite the non -significant finding for group differences in

PAGE 58

58 cluster size, there were also large effect sizes for between -group differences for TL and FL patients (d =. 88), and TL patients a nd controls (d =.73) with TL patients generating smaller clusters than both groups Data for clusters, switches, and mean cluster size across fluency type are pres ented in Figures 4 3 through 4 5 Additional Study Aims While not the primary aim s of our s tudy, we were interested in examining the relation ship amongst performances on naming analogues of our fluency paradigms and between naming and fluency performance Because scores on naming tests were not normally distributed, we used non -parametric test s (Kruskal -Wallis and planned Mann Whitney contrasts) t o assess for group differences in naming performance Number of actions correctly named on the Action Naming Test differed significantly by group ( H (2) = 19.54, p <.001), with controls correctly naming more actions than both TL patients ( U =21, z= 4.19, p <.001) and FL patients ( U =21, z= 2.74, p <.01). Contrary to our hypotheses, the two patient groups did not differ significantly in their ability to name actions (U =46, z= .225, p >.05). Performance on the F amous F ace N aming T est was assessed by computing a percent correct score, which was the total number of items correctly named out of the items correctly recognized, multiplied by 100. There were significant group differences in the ability to name famou s faces ( H (2) =18.95, p <.001). TL patients were able to accurately name only 33% of the faces they recognized, as compared to 53% for FL patients and 71% for controls. As expected, both TL and FL patients performed significantly worse than controls on this test (TL: U =22, z= 4.13, p <.001; FL: U =32.5, z= 2.07, p <.05) and consistent with our hypotheses, patients with TLE performed significantly worse than patients with FLE ( U =22.5, z= 1.97, p <.05).

PAGE 59

59 On a test of common object naming (B oston N aming T est ), cont rols were able to correctly name more items without cueing than both F L and TL patients ( H (2)=17.48, p <.001; TL: U =16, z= 3.97, p <.001; FL: U =15, z= 2.16, p <.05 ). On average, controls correctly named 55/60 items, FL patients named 49/60, and TL patients na med 43/60. Naming differences between FLE and TLE p atients were not statistically significant (U =18, z= 1.44, p >.05) but a moderate effect size (r =.33) suggests FL patients performed better than T L patients Spearman rho non -parametric correlations were conducted to examine the relationship amongst naming measures and between fluency and naming performance. When combining patients and controls, performance on the three naming tests was highly positively correlated. Performance on the ANT and BNT was also highly correlated with fluency test performance regardless of fluency type. Famous faces naming score was significantly correlated with scores on semantic fluency, name fluency, and action fluency. When we examined patients alone, BNT score was significant ly correlated with ANT (r =.542, p <.01) and FFNT (r =.508, p =.01) scores, but there was only a trend relationship between ANT and FFNT scores (r =.314, p =.08) Neither BNT nor FFNT score was significantly associated with performance on fluency measures. Perfo rmance on the ANT was significantly positively associated with phonemic fluency (r =.505, p =.01) and name fluency (r =.372, p <.05) scores. Correlations are presented in Tables 4 10 and 4 11.

PAGE 60

60 Table 4 1 Performance on fluency measures LTL FL Controls N 1 4 7 20 Phonemic (C) 10.1 (4.3) 9.60 ( 2.6) 13.3 ( 5.4 ) C+F+L 27.7 (11.3) 28.7 (4.7) 38.4 (11.1)* Semantic 18 .1 (5.4) 16.5 (4.2) 22.5 (6.5) Action 11.7 (4.5) 12.7 (2.9) 18.3 (5.8)* Name 6.1 (3.5) 8.8 (2.3) 14.5 (4.7)* Note: Means are presented, standa rd deviations in parenthesis. p < .01 for controls versus LTL and FL p <. 05 for controls vs LTL Table 4 2 Multivariate analysis of fluency performance in patients Pillais Trace F p Partial Eta Squared Phonemic + Action Model .015 0.1 4 .87 .015 Semantic + Name Model .280 3.54 .05 .282 Table 4 3 Four fluencies predicting patient group membership LTL (predicted) FL (predicted) % correct LTL (actual) 12 2 85.7 FL (actual) 4 3 42.9 71.4 Table 4 4 Semantic and name fluencies predicting patient group membership LTL (predicted) FL (predicted) % correct LTL (actual) 12 2 85.7 FL (actual) 2 5 71.4 81.0 Table 4 5 Correlations coefficients for fluency measures (patients only) (1) (2) (3) (4) (1) Phonemic Fluency 1.00 (2 ) Semantic Fluency .076 .74 1.00 (3) Name Fluency .144 .54 .415 .06 1.00 (4) Action Fluency .390 .08 .210 .36 .285 .21 1.00 Pearson correlation coefficients are presented followed by significance values.

PAGE 61

61 Table 4 6 Correlations co efficients for fluency measures (patients and controls) (1) (2) (3) (4) (1) Phonemic Fluency 1.00 (2) Semantic Fluency .388 .01* 1.00 (3) Name Fluency .628 .00 .407 .01* 1.00 (4) Action Fluency .637 .00* .446 .00* .662 .00* 1.00 Pearson correlation coefficients are presented followed by significance values. p < .01 Table 4 7 Correlations amongst neuropsychological measures (pati ents only) DS TMT B TMT B Err. WCST Cat. WCST Pers. BNT LM I LM II Semantic .108 .64 311 .18 .424 .07 .034 .89 .095 .73 .143 .78 .416 .08 .467 .05* Action .365 .11 .453 .04* .285 .24 .448 .07 .363 .17 .543 .02* .326 .19 .476 .04* Name .103 .66 .066 .78 .157 .52 .171 .51 .077 .78 .124 .62 .233 .35 .240 .33 Phonemic .494 .02* .5 10 .02* .282 .24 .231 .37 .408 .12 .415 .08 .29 0 .24 .095 .70 DS=Digit Span total score ; TMT B=Trail Making Test B time ; WCST=Wiscon sin Card Sorting Test ; BNT=Boston Naming Test, LM=Logical Memory Pearson correlation coefficients are presented followed by significance values. p < .0 5

PAGE 62

62 Table 4 8. Correlations amongst neuropsychological measures (patients & controls) DS TMT B TMT B Err. WCST Cat. WCST Pers. BNT LM I LM II Semantic .194 .243 .382 .028 .455 .01* .191 .27 .212 .23 .449 .01 .542 .00 .6 01 .00 Action .493 .00* .485 .00* .260 .13 .429 .01* .381 .02* .683 .00* .638 .00* .602 .00* Name .407 .01* .253 .13 .224 .20 .161 .36 .242 .17 .582 .00 .624 .00* .638 .00* Phonemic .527 .00* .430 .01 .285 .1 0 .222 .2 1 .392 .02* .566 .00* .328 .05* .453 .01* DS=Digit Span total score ; TMT B=Trail Making Test B; WCST=Wiscon sin Card Sorting Test ; BNT=Boston Naming Test, LM=Logical Memory Pearson correlation coefficients are presented followed by significance values. p < .01 p < .05 Table 4 9. Performance on naming measures LTL FL Controls BNT 43.4 (6.7) 48.6 (5.0) 54.9 (4.7) a ANT 47.4 (3.7) 48.0 (4.1) 53.1 (1.9) a FFNT 33.3 (21.1) 53.9 (20.4) b 71.9 (16.9) a Note: Means are presented, standard deviations in pare nthesis for ANT and BNT. FFNT is presented as % correct x 100. a Controls >TL and FL b FL>TL Table 4 10. Correlations between naming and fluency measures (patients and controls) FFNT ANT BNT Phonemic Fluency Semantic Fluency Name Fluency Action Fluenc y FFNT 1.00 .651 .00 664 .0 0 234 .07 .368 01* .433 00* .350 01* ANT 651 .0 0 1.00 .853 .0 0 413 .00 513 .00 565 .0 0 489 00* BNT .664 .0 0 853 .0 0 1.00 .403 01* 456 00* 596 00* 628 .0 0 Pearson correlation coefficients are presented followed by significance values. p < .01

PAGE 63

63 Table 4 11. Correlations between naming and fluency measures (patients only) FFNT ANT BNT Phonemic Fluency Semantic Fluency Name Fluency Action Fluency FFNT 1.00 .314 .08 .508 .01 ** .136 27 .177 .2 2 .003 .49 .116 .30 ANT .314 .08 1.00 .542 .01 ** .505 .01 ** .283 .10 .372 .04 .191 .20 BNT .508 .01 ** .542 .01 ** 1.00 .265 .14 .056 .41 .013 .47 .363 .06 Pearson correlation coefficients are presented followed by significance values. ** p < .01 p < 05 0 5 10 15 20 25 Semantic Phonemic Action Name Total number of words LTL FL Controls Control > LTL and FL at p<.05 Figure 4 1. Overall fluency performance across groups

PAGE 64

64 Figure 4 2. R eceiver o perating c haracteristic (ROC) curve for semantic and name fluencies predicting patient group

PAGE 65

65 0 0.5 1 1.5 2 Phonemic Semantic Action Name LTL FL Control significant at p<. 05 Figure 4 3. Mean number of clusters by group 0 4 8 12 16 20 Phonemic Semantic Action Name LTL FL Control significant at p< .05 Figure 4 4. Mean number of switches by group

PAGE 66

66 0 0.5 1 1.5 2 2.5 3 3.5 Phonemic Semantic Action Name LTL FL Control Figu re 4 5. Mean cluster size by group

PAGE 67

67 CHAPTER 5 DISCUSSION Summary of Findings The current study was undertaken to help understand differences in cognitive impairment in patients with left temporal lobe epilepsy versus epilepsy localized to the frontal l obes. More specifically, we sought to elucidate a distinct pattern of fluency test performance that would discriminate between these two patients groups We were interested in examining differences in both traditional semantic and phonemic fluency performa nce, as well as performance on two experimental fluency tests, action and name fluency. S cientific literature positing different neural networks for retrieval of action words and proper names suggested that incorporation of this material -specific content i nto traditional fluency test paradigms would improve measurement of the unique cognitive deficits associated with localized neural dysfunction. Furthermore, we hoped that an examination of fluency strategy, including generation of clusters, cluster size, a nd switches between clusters, would provide useful diagnostic information about our patients. In other words, by manipulating the fluency retrieval demands involved and examining the cognitive strategies employed, we hope d to more accurately discriminate b etween patient groups, in addition to advancing our understanding about the neural specificity of the brain regions involved Th e first primary aim of this study was to characterize performance of patients with frontal or left temporal lobe epilepsy, and matched healthy controls on a panel of verbal fluency tests that included clinical measures of semantic and phonemic fluency, and experimental measures of action and proper name fluency. We hypothesized that patients with epilepsy would perform worse on a ll of these measures than would our healthy controls. We also believed that patients with temporal lobe epilepsy would evidence impaired semantic fluency (but not phonemic

PAGE 68

68 fluency), while patients with frontal lobe epilepsy would demonstrate impairments on phonemic fluency (but not semantic fluency). With regard to our experimental fluency measures, we believed that overall fluency score on a ct ion and proper noun fluency would doubly dissociate patients with frontal and temporal lobe epilepsy, with frontal lobe patients performing worse on action fluency and temporal lobe patients exhibiting comparative deficits on tests of proper name fluency. These hypotheses were only partially confirmed. As expected, controls generated more words across fluency test s tha n both patient groups. Interestingly, however, the findings were statistically significant only for the experimental action and name fluenc y tests, with controls outperforming patients with both types of epilepsy. While there were trends for signif icance a nd moderate effect sizes for patient and control group differences on semantic and phonemic fluency these effects did not reach a level of statistical significance. Contrary to our predictions there were no statistically significant differences between t he two epilepsy groups on any of the fluency measures. However, there was a moderate effect size for name fluency, with TL patients generating fewer proper names than F L patients suggesting meaningful group differences that could not be adequately detecte d because of our sample size In both patient groups, participants generated the most words for supermarket fluency, follow ed by action, phonemic, and name fluency a finding consistent with past literature (Piatt et al., 1999) Our second aim was to dete rmine the predictive validity of traditional and experimental f luency measures in predicting patient group membership We predicted that both traditional and experimental fluency measures would adequately discriminate between patients with frontal and temp oral lobe epilepsy, and that the multivariate combination of our four fluency measures would have superior predictive abilit y above and beyond either traditional or experimental fluency measures alone. These hypotheses were not confirmed in our study popul ation. We

PAGE 69

69 found that our fluency panel did an inadequate job of accurately predicting patient group membership. Our four -fluency model better predicted temporal lobe than frontal lobe involvement, but was still a poor fit for our data Our final model, whi ch fit the data well, included only semantic and name fluency and accurately predicted 86% and 71% of our TL and FL patients respectively This suggests that the tests removed from the model ( action and phonemic fluency) did not offer additional predictiv e value beyond the variables in our final two predictor model. The third aim was to examine the conve rgent and discriminant validity of our experimental fluency measures using traditional neuropsychological measures of frontal and temporal lobe functioni ng as criterion variables We hypothesized that performance on tests of c om mon and proper noun fluency would be more related to measures also sensitive to the integrity of the temporal lobe (i.e., related to semantic stores) while action and phonemic fl uency scores w ould exhibit small to moderate relationships with traditional measures of executive function When examining performance for controls and patients combined, we found strong positive correlations amongst all four fluency measures, suggesting the presence of a common source of variance in fluency performance regardless of the retrieval demands. When only patients were included in the analysis, phonemic and action fluency were moderately positively correlated, as were name and semantic fluency. Name and semantic fluency were uncorrelated with both action and phonemic fluency, suggesting that in patients with localized neuronal dysfunction, fluency performance was differentially impaired based on the category -specificity of the material to be retr ieved. When we examined relationships between our fluency measures and other measures of neuropsychological functioning for our patients and controls, there were significant positive

PAGE 70

70 correlations between all four measures and scores on tests of language and verbal memory ( BNT, LM I and LM II) This was consistent with our predictions for name and semantic fluency, but not action and phonemic fluency. As predicted, action and phonemic fluency were correlated with measures of executive function (Digit Span number correct, TMT B time and WCST perseverations ). Contrary to our prediction, semantic fluency was correlated with performance on TMT B (time and errors) and name fluency was related to performance on D igit S pan. The four th aim of the present study wa s to determine whether a qualitative analysis of fluency performance, including number of clusters, switches, and cluster size, would dissociate performance of patients with FLE, TLE, and healthy controls. Based on existing literature, w e hypothesized that patients with TLE would exhibit reduced cluster size, particularly on tests that carry a heavier semantic burden ( name fluency, semantic fluency) and that patients with FLE would evidence a reduced number of switches, pr imarily on tests of phonemic and a ction fluency. These hypotheses were only partially confirmed. There were no significant group differences for clusters, switches, or cluster size for phonemic or semantic fluencies. On action fluency clusters, controls generated more clusters than did eit her patient group, though this was statistically significant for only for the control versus TL comparison. Controls also switched more frequently than TL patients. There were no significant group differences in action fluency cluster size. On name fluenc y, TL patients generated significantly fewer name clusters than controls, and both TL and FL patients switched less frequently than controls. T here was a large effect indicating TL patients also generated fewer clusters than FL patients. Large e ffect sizes also revealed that TL patients generated smaller clusters than FL and controls on name fluency, which is consistent with our a priori hypothesis.

PAGE 71

71 The body of naming literature posits distinct neural substrates for action naming (i.e. the frontal cortices ) and person naming (i.e. the anterior temporal cortices of the left hemisphere), and while there is preliminary support for a comparable fluency substrate, these paradigm have not been used extensively in the literature. Because of this, we were interest ed in examining the relationship between our fluency measures and related naming measures, including the Boston Naming Test, Action Naming Test, and Famous Faces Naming Test. We hypothesized TL patients would exhibit more prominent naming deficits than FL patients and controls on measures of common object naming and famous faces naming, but that patients with FLE would exhibit deficits on our measure of action naming. Our naming hypotheses were also only partially confirmed. As expected, controls were able to correctly name more items on all three naming measures than both FL and TL patients. On the BNT, no statistical differences were found between patient groups, but a moderate effect size suggested that TL patients were more impaired at naming common obje cts than FL patients. TL patients were also worse at naming familiar famous faces as compared to patients with FL epilepsy; in fact, they named 20% less than patients in the FL group. Contrary to our predicted results, FL and TL patients did not differ in their ability to name actions on the ANT. When examining the relationship amongst naming measures, performance was highly correlated for patient and controls. Performance on the ANT, BNT and FFNT was also highly correlated with fluency test performance re gardless of fluency type except for FFNT and phonemic fluency When we examined patients alone, BNT score was significantly correlated with ANT and FFNT but no significant relationship was established between ANT and FFNT Neither BNT nor FFNT score was significantly associated with performance on fluency measures. Performance on the ANT was associated with phonemic and name fluencies

PAGE 72

72 In sum, while many of our hypotheses were confirmed in the present study, a significant p ortion were also disconfirmed, p articularly as it relates to performance on traditional measures of fluency, action fluency, and our qualitative analysis of fluency performance I nterpretation s of our study findings are presented below. Interpretation of Findings Semantic and Phonemic Fl uenc y We found that our patient groups did not show differential phonemic and semantic fluency performance and that neither measure was a good predictor of patient group membership. These finding s were inconsistent with our hypotheses but not entirely sur prising and part of the reason we undertook the present study comparing our experimental tests to these traditional measures While there are many studies that show successful semantic fluency performance relies more on the integrity of the left temporal lobe and phonemic fluency is sensitive to the presence of frontal lobe pathology, there are numerous studies that fail to show this effect. Thus, the current results are in good company. Studies have shown equivalent test performance on semantic fluency in patients with ante rior and posterior lesions. In an epilepsy population, Drane et al. (2006) found that patients with frontal lobe seizure foci were more impaired than a group with temporal lobe epilepsy on measures of semantic fluency, contrary to thei r hypotheses that pr edicted more impairment in the temporal lobe group. Another study comparing patients with focal anterior and posterior lesions found that both types of lesions produced impairments on semantic, or category, fluency (Stuss et al., 1998). Additional studies employing a variety of populations have found a similar pattern of equivalently impaired semantic fluency in frontal and temporal lobe patients (Baldo & Shimamura, 1998; Costello & Warrington, 1989; Owen et al., 1990; Randolph et al., 1993). These authors have argued that rapid word generation, regardless of retrieval demand can be

PAGE 73

73 impaired by frontal lobe lesions. These studies purport that patients with executive dysfunction are unable to perform effective, strategic, and efficient se arches for words, irrespective of whether the search is semantically or phonemically driven (Baldo et al., 2006; Baldo & Shimamura, 1998; Troyer et al., 1998) A similar pattern has also been established for phonemic fluency tests (Stuss et al., 1998; Mill er, 1984; Pendelton et al., 1982; Perret, 1974). Emory and Alvarez (2006) found that the bulk of studies in their meta analysis of frontal lobe lesion patients reported significantly poorer phonemic fluency scores compared to controls, however, a signific ant percentage found similar impairment on phonemic fluency in patients with non-frontal lobe lesions. Henry and Crawford (2006) found that phonemic fluency deficits were largest in patients with left frontal lesions, but that patients with non -frontal le ft hemisphere lesions were often similarly impaired, suggesting phonemic fluency performance may be determine d both by an executive factor and a verbal component. In fact, this theory was put forth decades ago by Ramier and Hcaen (1970), who hypothesized that successful performance on phonemic fluency is determined by an executive factor located within the frontal lobes and a verbal factor mediated more generally by the language -dominant hemisphere. Equivalently i mpaired performance on semantic and p honemic fluency in patients with frontal and temporal lesions suggest s that these tests are sensitive to the presence of frontotemporal damage, but not specific to more localized impairment within this region. Adequate performance on semantic and phonemic fluency tasks is likely multi -factorial, and may depend on verbal contributions from the language dominant hemisphere, efficient search and retrieval strategies dependent on frontal lobe functioning and a general cognitive factor, or g This was reflect ed in our study ; semantic and phonemic fluency performance was significantly

PAGE 74

74 correlated with measures of verbal/ semantic ability but was also related to measures thought to tap an executive component Further, WASI Full Scale IQ was a strong predictor of f luency score (R2=.26, and .30 for semantic and phonemic fluency) suggesting that fluency performance, regardless of the retrieval demands, could also be nonspecifically depressed in both groups due to an overall cognitive impairme nt associated with chronic uncontrolled seizures (Jokeit & Ebner, 2002) That being said, intact verbal and semantic memory abilities and efficient search/retrieval strategies likely contribute in different ways to performance on these measures, with the former being more importa nt to semantic fluency performance and the latter to phonemic fluency (Butters et al., 1987; Gleissner & Elger, 2000; Janowsky et al., 1989; Jurado, et al., 2000; Martin, Loring, Meador, & Lee, 1990; Monsch et al., 1992; NKaoua, 2001; Rosser & Hodges, 1994; Stuss et al., 2000; Troster et al., 1995; Troyer et al., 1998). This imp lies that the original hypothese s pertaining to our patient groups may be valid, but due to the nature of our study population, were unable to be borne our within the constraints of our current study. Power analyses based on data from Troyer and colleagues (1998) suggested that between group differences could be detected on semantic and phonemic fluency with a sample of sevento ten patients with leftlateralized frontal and temporal lesions. While studies do not provide conclusive evidence that phonemic and semantic fluency performance can be doublydissociated, the discrepancy between our findings and those projected are most likely due to differences in our sample populations. Ma ny studies that have found group differences in TL versus FL populations, including Troyer et al. (1998), included patients with circumscribed lesions from stroke and tumors, in addition to etiologies such as traumatic brain injury and seizure surgery. Sim ilar to our study, her study included a mixed population of relatively acute (i.e., post -stroke)

PAGE 75

75 and chronic patients (i.e., post -surgical intractable epilepsy). Contrary to ours, however, her patients all had stable (> 3months post injury) and focal (dors olateral prefrontal cortex; superior medial frontal; inferior medial frontal) lesions. Our study population was different from this in many regards. A pre -surgical epilepsy population likely suffers from both diffuse (e.g., a history of intractable seizure s with propagation to surrounding brain regions ) and focal (e.g., localized onset) impairments and can experience overall cognitive depression due to epilepsy medications or the duration of their disease ( Jokeit & Ebner, 2002; Loring, Marino, & Meador, 2007; Nichols, Meador & Loring, 1993). Additionally, m ost of the pre -surgical patients in our study were actively experiencing seizures whic h contributed to the chronicity and pervasiveness of their cognitive dysfunction, making their pattern of neuronal da mage dramatically different that patients who experience an acute injury such as a stroke. Furthermore, due to recruitment constraints, our study included both pre and post -surgical epilepsy patients and patients with both right and left frontal lesions, both of which are factors that could explain our lack of predicted group differences. By including both pre and post surgical patients, some of whom had circumscribed, defined, and stable lesions and some of whom did not, we introduced more variance into our study sample. The inability to include only left frontal patients in our study was possibly the single biggest explanatory factor for our non significant findings. The left frontal lobe, particularly the dorsolateral prefrontal region (DLPFC), has bee n shown to be the most critical region to phonemic fluency performance (Milner, 1964; Pendelton et al., 1982; Perret, 1974; Stuss & Levine, 1998), while the right DLPFC appears to be a less important region. Patients with right DLPFC damage ( Miceli et al., 1981; Miller, 1984; Ramier & Hacean 1970; Troyer et al., 1998) display impaired phonemic fluency, but to a lesser degree than their left hemisphere counterparts. There is some evidence

PAGE 76

76 that the right DLPFC contributes to on -task behavior important to f luency performance, such as monitoring, retrieval success, and inhibition of extraneous information (Cabeza and Nyberg, 2001) However, the left dorsolateral prefrontal regions are also thought to play a key role in the se same processes in addition, perhaps, to its preferred access to the lexicon Imaging data supports primary involvement of the left DLPFC, and secondary involvement of right frontal structures, in phonemic fluency performance (Frith et al., 1995; Parks et al., 1988). Damage to the superio r medial frontal regions in both the right and left hemisphere can impair fluency performance (Stuss & Levine, 2002; Troyer et al., 1998) but are again thought to play a secondary role to the DLPF C. Unfortunately, the nature of our patient group did not al low for more precise localization of pathology within particular sectors of the frontal lobe. However, b ecause of our mixed sample of left and right frontal lobe epilepsy patients, the demands of phonemic fluency tests (i.e., search and retrieval, organiz ation of unstructured orthographic information, flexibility) may not have been as taxing as they might have been in a more pure sample of left -lateralized patients, thereby reducing the extent to which impairments were found. Action Fluency Action fluenc y is a relatively new test construct that grew out of the naming literature in agrammatic aphasics showing impaired retrieval of words denoting actions in the presence of spared object naming (Miceli, 1984). The lesion literature also supports deficits in action naming associated with damage to the frontal cortices. For instance, Damasio & Tranel (1993) demonstrated a double dissociation in the performance of patients with anterior temporal cortex damage ( who had difficulty naming pictures of objects) compa red to another patient with left premotor damage ( who was unable to name actions depicted in line drawings ). Deficits in action naming have been inconsistently demonstrated in patients with fronto -temporal dementia and

PAGE 77

77 various other lesions (Damasio & Tran el, 1993; Monsch et al., 1992; Ostberg et al., 2005; Silveri et al., 2003). Action fluency, on the other hand, has been relatively unstudied until recently. Preliminary stud ies with HIV and Parkinsons dementia patients have demonstrated impairments in ac tion fluency compared to semantic and phonemic fluency. These studies have also supported the idea that action fluency is a construct of executive functioning ( Piatt et al., 1999a and 199b; Woods et al., 2005a and 2005b ). These findings were only partially replicated in our patients with frontal epilepsy. We found that frontal and temporal lobe epilepsy patients performed similarly on both action word naming and action fluency. As a result, action fluency score was a poor predictor of patient group member ship. We did find support for the notion that, at best, action fluency may indeed be a construct sensitive to executive functioning, or that it at least had some relationship with other measures purported to be sensitive to executive function. Action flu ency, more so than phonemic fluency, was strongly related to performance on the WCST Categories, WC ST Perseveration, Digit Span, and Trails B time, providing evidence of convergent validity amongst measures of executive functioning. Also noteworthy, howeve r, is the fact that action fluency scores were moderately correlated with measures of verbal/semantic ability, including the BNT and WMS -LM II. This has not been found previously in the literature, but has also not been explored fully as Woods et al. ( 2005) did not include measures of naming or verbal retrieval in their analysis of correlates of action fluency. Our findings, in conjunction with previous findings of Woods and Piatt, indicate that adequate performance on this test is multi determined, relate d both to executive functioning abilities and to verbal abilities. Neuroimaging studies also support the notion that both anterior (frontal operculum, left premotor, left prefrontal, left insula) and posterior (left mesial occipital cortex, left supramargi nal and po sterior

PAGE 78

78 temporal regions) cortices may play a role in the retrieval of words denoting actions (Damasio et al., 2001; Tranel et al., 2001). Impaired action fluency performance in our temporal lobe group may be in part, attributed to this factor. A nother possible explanatory factor is that temporal lobe epilepsy patients commonly exhibit difficulty with components of executive functioning (response dis inhibition, impulsivity, set loss, and difficulties with mental fle xibility and abstract thinking ) secondary to propagation of seizure related neural noise from temporal to frontal regions via the medial and lateral limbic circuits (Hermann & Seidenberg, 1995) Moreover, a growing body of literature suggests that structural and functional abnormaliti es in TLE patients exist not only within TL structures, but also in regions outside of the temporal lobes. For instance, significant white matter changes have been demonstrated in extratemporal cortex, including the frontal lobes (Hermann et al., 2003; Oye gbile et al., 2006). This pattern of impaired executive functions has been well documented on the WCST, TMT and Stroop paradigms amongst others (Hermann, Wyler, and Richey, 1988; Martin et al., 2000; McDonald et al., 2005; Trennery & Ja ck, 1994; Corcoran & Upton, 1993). In fact, many of these studies have found patients with language dominant temporal lobe epilepsy to be equally or even more impaired than those with frontally mediated seizures. This explanation has also been used in part to explain simila r ly impaired performance on phonemic fluency tests, and may also extend to our tests of action fluency. Again, the possibility remains that our temporal lobe patients with longstanding seizure disorders may also have exhibited depressed cognitive profiles on multiple cognitive domains due to the cumulative effect of uncontrolled seizures (Jokeit & Ebner, 2002). Additional differences in our study population and those of Woods and Piatt may help to explain our discrepant findings. The action fluency construct has been used only in studies of

PAGE 79

79 patients with frontal and subcortical disease, namely HIV and Parkinsons Disease. Both populations can exhibit significant executive dysfunction, but also deficits in motor and cognitive processing speed which are not necessarily hallmarks of focal seizure disorders The impact of cognitive slowing on fluency performance was not accounted for in studies by Piatt or Woods, despite significant relationships between action fluency scores and scores on measures of cognitive and motor processing (Woods et al., 2005) It may be the case that the combin ation of executive dysfunction and cognitive slowing in these populations differentially affected action fluency performance compared to semantic fluency, for instance, which cou ld help explain the intact performance of our frontal lobe epilepsy patients This could be the case for action fluency in particular given the hypothesized executive burden of the test. Finally, as with semantic and phonemic fluency performance, our mixed sample of left and right frontal patients probably reduced our ability to find a significant effect that m ay have otherwise been present in a solely left -frontal sample. With regard to the two predominant theories that exist to explain the discrepancy between retrieval of action versus object words, the first states that knowledge about objects and actions is stored in association cortices adjacent to the primary cortical regions that process these classes of stimuli (Damasio & Tranel, 1993; Perani et a l., 2009). As such, object knowledge is stored in cortical regions adjacent to the occipito temporal visual stream, while action knowledge is stored adjacent to structures in the frontal lobe including the prefrontal cortex, premotor cortex, and supplement ary motor area and focal lesions to these areas can disrupt successful retrieval of action words. The second theory holds that the deficit is largely executive in nature, and relates to the difficulty of mentally coordinating and manipulating the large am ount of information related to action -words ( de Nbrega, Nieto, Barroso, & Montn, 2007; Grossman, 1998; Silveri

PAGE 80

80 et al., 2003; White Devine et al., 1996). This latter theory also posits that as with all fluency paradigms, verbal ability may mediate over all fluency score regar d less of executive capacity. In reality, the se theories are not mutually exclusive and the scientific literature provides support for both. It is likely the case that while acute lesions to premoto r or association motor cortices dis rupt successful retrieval of action words, insult to other neuronal regions or pathways is also sufficient to impair this ability. The current study provides greater support for the second theory, though the former cannot be tested fully because our group was not comprised of patients with focal lesions in the aforementioned regions. Name Fluency Generally speaking, our findings with regard to performance on tests of proper n ame generation were consistent with our hypotheses. Name fluency proved to be the most demanding fluency measure for all patients and controls, but TL patients were differentially impaired on this test, suggesting a true deficit in this ability rather than a main effect of task difficulty. Patients with lefttemporal lobe epilepsy, both pre and post -surgical, showed the weakest performance on this measure, generating on average only six accurate responses in the span of a minute, which was statistically different than controls and different from FL patients based on measures of effect size. Indeed, name fluency was the strongest predictor of group membership in our regression analysis. In fact, classification accuracy statistics for name fluency alone were rather comparable to the model that additionally included semantic fluency with the two -predictor model correctly classifying only two additional TL patients. As we expected, in our patient groups, name fluency was related to the traditional measure of semantic fluency, but not to measures of action or phonemic fluency, providing evi dence for the convergent and discriminant validity of the measure as closely allied with semantic retrieval When we examined patients and controls combined, our correlations show that this measure is

PAGE 81

81 strongly related to other tests assessing verbal and/or semantic abilities such as the BNT, LMI and LMII reflecting good external validity for a measure thought to be sensitive to temporal lobe functioning. H owever, for patients alone, no significant relationships with measures of language or memory both lar gely mediated by language dominant temporal structures were seen While surprising initially, this finding is consistent with studies showing that deficits on famous face naming tests may be dissociated from deficits on measures of common object naming ( Drane et al., 2008). In other words, this test construct may be mediated less by general verbal abilities than most fluency tests, and may rely more upon a unique semantic factor not tapped by other assessment instruments This feature may make famous f ace naming particularly sensitive and specific to anterior temporal lobe functioning A roughly equivalent pattern of results was exhibited when we embedded this test construct in a confrontation naming paradigm, lending further support to our hypothesis that generation of proper names is dependent on the integrity of temporal lobe structures. For patients and controls, significant relationships were found amongst the FFNT and all other fluency and naming tests, but when we examined patients alone the re lationship with the BNT emerged as the only significant relationship, followed by the weak correlation with the ANT. On this famous faces naming measure, temporal lobe epilepsy patients were able to correctly generate names for only 1/3rd of the familiar f aces compared to roughly and for t he frontal lobe patients and controls, respectively. Our findings provide support for the notion that the anterior portion of the left temporal lobe plays a critical role in the ability to generate names or apply lab els to people or objects and is particularly important in the case of proper names. Existing lesion studies (Fukatsu et al., 1999; Glosser, Salvucci, & Chiaravalloti, 2003; Martins & Farrajota, 2007) and functional imaging data ( Gorno Tempini et al., 1998; Grabowski et al., 2001; Tranel, 2006; Tranel,

PAGE 82

82 Grabowski, Lyon, & Damasio, 2005; Tsukiura et al., 2002) provide support for this hypothesis. This area has been deemed a convergence zone by Damasio, Tranel, and colleagues (Damasio, et al., 1996; Damasio et al., 2004; Tranel, Damasio, & Damasio, 1997;Tranel et al., 2003) These authors propose that the anterior temporal lobe serves as a region that helps bind multi modal sensory and motor input from our surroundings, leading to the development of amodal c oncepts. Since names are arbitrary labels that denote members of conceptual categories, damage to this area can produce category -specific deficits in word retrieval. Specific defi cits have been shown for animals and unique entities such as people and landmarks (Damasio, 1996; Fukatsu et al., 2000; Glosser, Salvucci, & Chiaravalloti, 2003; Gorno Tempini et al., 1998; Grabowski et al., 2001; Milders, 2000; Tranel, 2006). The apparent difficulty with retrieval of the latter has to do with the semantic un iqueness of the object or person (Semenza & Zettin, 1989; Glosser, Salvucci, & Chiaravalloti, 2003; Grabowski et al., 2001; Tranel, 2006). Whereas common names refer to concepts, or a set of attributes that are shared by multiple entities within the same concept, proper names do not inherently contain attributes in and of themselves and are merely expressions by which we refer to an individual person or item. Because of this, it is thought that widespread neural networks support the representation of comm on nouns, while proper nouns are thought to hold rather fragile associations with their unique reference ( Gorno Tempini & Price, 2001; Martins & Farrajota, 2007). This distinction is particularly salient when contrasting between generation of common vers u s proper names on fluency and naming tests and may explain why the latter are emerging as particularly sensitive to anterior temporal lobe damage For instance, the semantic representation for the word dog (an appropriate response for the semantic flu ency category Animals) is l ikely much more substantial perhaps encompassing the words beagle, pug,

PAGE 83

83 dalmation, poodle, and the like, whereas Bob Dole, Marilyn Manson, Hulk Hogan, and Mother Theresa all refer to singular, unique entities, not linked in cohesive semantic networks Our findings suggest that a name fluency paradigm may offer particular value in detecting the type of impairment that is common in both pre and post -surgical language -dominant temporal lobe epilepsy patients. Na me fluency score appears to be less affected by frontal lobe fluency processes such as efficient monitoring/searching/flexibility than are other measures of fluency, including semantic fluency, and are more contingent upon adequate functioning of the temporal lobe semantic networks. It may also be the case that a name fluency paradigm could reveal impairments that are not evident on other types of neuropsychological tests Drane and colleagues (2008) have described patients with subjective complaints of pos t -surgical naming deficits who perform at expectation on measures of common object naming but show impairment on their famous -faces naming test This suggests that traditional clinical measures, which focus only on object naming, may not adequately tap th e type of abilities commonly disrupted by TLE or anterior temporal lobectomy. Qualitative Analysis of Fluency Performance Unfortunately, we did not find that an analysis of qualitative fluency performance provided much additional useful information about the patient groups in our study. In general, the pattern of qualitative fluency performance echoed the quantitative analysis of performance; no significant group differences were found on measures of semantic and phonemic fluency. TL patients generated fe wer clusters and switches on action fluency than controls and fewer and smaller clusters than controls and FL patients on name fluency. Both patient groups switched less frequently than controls on name fluency; all results that mirror the quantitative im pairments just discussed.

PAGE 84

84 This raises the following question : are number of clusters, cluster size, and switches actually proxies for overall fluency performance or are they independent measures of the functional integrity of the temporal and frontal str uctures ? There is evidence to support the former (proxy) view, as qualitative measures of fluency have been shown to be highly tied to overall fluency score in an Alzheimers and Parkinsons sample (Troyer et al., 1997). This pattern was replicated in our data ; measures of qualitative performance, across group and fluency task, were correlated with overall number of words generated. Most studies of clustering and switching as measures of temporal and frontal lobe functioning have found a similar pattern (Troyer et al., 1998a, 1998b; NKaoua et al., 2001) When s tudies have found overall group differences in fluency score, they also have reported differences in clustering and switching across tasks. When group differences were not found for overall total scor e, differences did not tend to emerge on clustering and switching analyses. Troyer addresses this issue directly (1998a), and argues against this point, but does recogniz e that switching score and words generated may be correlated variables because the nu mber of words generated were always associated with group differences in switching. In earlier work, s he provide s direct evidence that decreased clustering does lead to an overall decrease in total words produced (Troyer et al., 2007). It is certainly pl ausible that two patients with the same overall fluency score of ten, for instance, could produce very different patterns of performance (i.e., orange grape -apple -cheeseyogurt -butter -paper towel -napkin-toilet paper versus orange -hot dog-paper towel -pencil s wine beer -champagne -cat food -cheese-carrots ), thereby achieving differing sco res on clustering and switching. However, the data from our study and others finds this scenario less likely when differences in overall score are not apparent.

PAGE 85

85 An alternative explanation is that clusters and switches are not necessarily proxies of overall fluency score, but that clusters/switches and overall score are both proxies of temporal and fronta l lobe functioning Under this premise, the lack of significant findings across most of qualitative (and quantitative) measures has to do with the underlying level of localized neuronal dysfunction in our patient groups. Again, the heterogeneous nature of our sample, our frontal lobe group in particular, may have lessened our abil ity to replicate the effects found in other studies of patients with m ore circumscribed lesions of the left or bilateral frontal lobes. Troyer and colleagues (1998a) were able to compare performance across subgroups of frontal patients (i.e., left dorsolateral prefrontal cortex (LDLPFC), right dorsolateral prefrontal cortex (RDLPFC), superior medial frontal (SMF) cortex, inferior medial frontal (IMF) cortex) and found between-group differences that support this assertion; specifically, switching was impaire d in the LD LPF and c ombined SMF groups, but not the RDLPF and IMF groups. This finding is consistent with other neuroimaging and cognitive studies showing impaired initiation of behavior, poor cognitive fl exibility, and perseverations are most strongly rel ated to the left dorsolateral frontal, inferior frontal, and anterior cingulate regions (Hirschorn & Thompson Schill, 2006; Troster et al., 1998). Most likely, our sample included patients whose damage transcended these functional boundaries muddying any profile that may have resembled those previously reported with regard to switches in particular. Despite our non -significant findings between patient groups with regard to switches, and clusters on action, phonemic and semantic fluency, the reduced clust er number and size on name fluency in our temporal lobe group is an interesting finding, and consistent with our a priori hypothesis. Unfortunately, our small sample size kept these numbers from reaching statistical significance, but our large effect sizes suggest they are indeed meaningful findings Our results

PAGE 86

86 show that not only did patients with left temporal lobe epilepsy have difficulty generating proper names, they also had more difficulty than frontal lobe patients or controls in linking response ite ms to each other in a semantically meaningful way. When they were able to successively generate semantically related names, they tended to be able to generate fewer of these names before exhausting the semantic network in which they reside These findings on name fluency are particularly interesting, as this was the measure we hypothesized to be most sensitive to impairment of semantic networks subserved within anterior temporal structures. Consistent with our hypotheses pertaining to overall name fluency performance, this reduced ability to generate semantically related proper names in particular may reflect the disruption or degradation of semantic memory stores within the temporal lobes. The finding of reduced clusters and cluster size on tests sensitive to temporal lobe functioning has been reported frequently; Troyer has hypothesized that the best indices for discriminating patient groups were phonemic -fluency switchingand semantic -fluency clustering (Troyer et al., 1998a, 2000; Reverberi, Laia cona, and Capitani, 2006). This observation indicates that a combination of overall fluency score and an analysis of cluster ing and switch ing by fluency type may provide useful information about the underlying cognitive impairments in various patient groups. Limitations of the Present Study There were a number of factors that limited the present study and may have affected our ability to find the results we predicted. We have previously discussed most of these, but they will be reviewed herein. First and forem ost, our small heterogeneous sample likely negatively impacted our study. Unfortunately, even with an extended recruitment time of eighteen months, we were unable to collect data on enough pre -surgical patients with left lateralized temporal and frontal lo be epilepsy. This may have been due to a number of factors, though the primary reason was decreased patient flow through Shands inpatient epilepsy monitoring unit. To supplement

PAGE 87

87 patient flow, we concurrently recruited post -surgical epilepsy patients who ha d undergone surgery between 2000 and 2007. By employing this recruitment strategy, we were able to meet our expected sample size for our left temporal group. However, we were still not able to recruit enough participants with left frontal epilepsy and in f act, no addition left frontal patients were able to be recruited through post -surgical mailings. The alternative to this solution was to include patients with frontally -localized epilepsy, regardless of laterality. Through this means, we were able to meet our projected sample size in both patient groups. However, this likely limited our ability to test out our hypotheses as they pertained to left frontal lobe functioning in particular. Many of our predictions with regard to test s of frontal lobe functioning (i.e., phonemic and action fluency) were based on the notion that the left frontal region, the DLPFC specifically, plays a key role in successful fluency performance. This could not adequately be explored in our current sample because of the mixed nature of our group. While other areas of the frontal lobes, including the bilateral portions of the superior medial frontal lobe may also be involved in fluency performance, the lack of specificity in our sample prohibited us from testing the contribution of va rious regions to task behavior. Nonetheless we did still find interesting differences related to some of our study hypotheses largely related to tests sensitive to temporal -lobe functioning, which were less impacted by the heterogeneous nature of this sa mple. Many of these findings still only reached clinical, not statistical si gnificance, implying our study m ay still have been slightly underpowered. That being said the pre -surgical patients we screened and recruited into our study were consecutive admi ssions to a clinical epilepsy center and are representative of the actual type of patients neuropsychologists are asked to assess for pre -surgical evaluations. Patients with epilepsy often do not have clearly localized seizure s They may have multiple seizure foci,

PAGE 88

88 seizures that propa gate from one brain region to another, a mixed pattern of focal and generalized events, or have a mixed profile of electrographic and non-epileptic seizures, all of which can complicate the localization/lateralization process. These patients also frequently present with psychiatric illness, past surgeries, significant head injury, and comorbid diagnoses of learning disability or mental retardation, and take multiple medications, all obscuring the diagnostic picture even farther So in reality, the sample with which we tested our measures and hypotheses was not ideal, but in many ways, most representative of the type of patient on which these measures would be used clinically. While t his limits our ability to make conclusions abo ut the sensitivity and validity of our measures, it provides useful information about how a typical patient with frontally -mediated seizures might perform. One of the initial points of the current study was to assess the utility of our standard fluency me asures and develop new measures more sensitive to the presence of frontal or temporal lobe dysfunction. This is particularly important as it relates to our ability to provide useful information to epileptologists and epilepsy neurosurgeons about laterality and localization of seizure onset based on cognitive test patterns. Though we can make stateme nts about the sensitivity and sp ecificity of these measures as they relate to frontal and temporal functioning in general we cannot decisively comment on their ability to predict seizure localization pre surgically. All of the patients in our final sample had medication refractory epilepsy, though a portion had undergone resective surgery to alleviate their seizures Cortical resections for epilepsy tend to be ra ther focal removing only the epileptogenic foci if possible, though often surrounding tissue may also be removed or compromised. For our post -surgical patients, surgical intervention tended to be curati ve, as the majority were seizure free fol lowing their resections This alleviation of seizures likely promoted overall brain health, but could have also introduced

PAGE 89

89 more cortical damage than was initially caused by the seizures themselves. This could have obscured the true cognitive picture that may have been present in a strictly pre -surgical population, making conclusions about performance in that population difficult. Unfortunately, our samples of pre and post -surgical patients were too small to conduct any meaningful analyses to compare group differences wi th regard to demographic, m edical, or cognitive variables for these patients. Finally, our small sample size prohibited us from examining other factors of interest that may have impacted our fluency and naming results. This includes an analysis of the rol e of age of seizure onset, the absence/presence of lesions, propagation of seizure activity, and effects of anti -epileptic medication. Directions for Future Research and Clinical Use Phonemic and semantic fluency measures are two of the most commonly use d tests by neuropsychologists. These tests have well established, demographically corrected norms, making them appropriate for Caucasians and African -Americans, and persons from their first through their ninth decade of life ( Delis, Kaplan & Kramer, 2001; Heaton, Miller, Taylor & Grant, 2004 ). These tests have been used to characterize the cognitive performance of virtually every type of patient, including those with dementia, epilepsy, TBI, infection, and tumor They obviously offer clinical value as part of neuropsychological assessments and will continue to be used in the future. The current study suggests that used individually, they may not offer definitive localizing value with regard to clinical epilepsy patients. As previously mentioned, the sample of epilepsy patients in this study is thought to be representative of the overall population of patients who present for evaluation in surgical epilepsy centers. In our sample, scores on tests of phonemic and semantic fluency were overlapping, and did not distinguish between patient groups, which is

PAGE 90

90 part of the usual question that motivates pre -surgical neuropsychological assessments. As such, configural interpretation of cognitive profiles should include fluency tests along with other tests sensitive to th e presence of localized neuronal dysfunction. Action fluency is a relatively new construct in the literature and has only been used in limited populations at the present time. These populations tend to have both frontal and subcortical disease involvement making it difficult to pinpoint the exact underlying process that is impairing action fluency performance. Furthermore, no definitive theory has emerged in the literature that is sufficient to explain the neural underpinnings supporting this construct to the exclusion of other viewpoints Complicating the theoretical debate is that fact that cognitive, neuroimaging, and lesion data exist that support both main theories, raising the possibility that they may not be mutually exclusive. Our findings do not provide significant clarity to this debate. To our knowledge, there have been no published studies that have examined the action fluency construct in a surgical epilepsy population. Based on the current findings there appears to be limited support for the use of an action fluency paradigm in clinical epilepsy at the present time To the extent that our sample represents a real -life population, the test appears to have little predictive power and its sensitivity to frontal lobe functioning remains question able. A significant amount of theory driven research needs to be done to show the specificity of th is test and its construct validity before it is used clinically. More specifically, a series of well -designed studies contrasting patients with circumscribe d premot or/supplementary motor and DLPFC lesions would shed light on the most important cortical regions subserving test performance. Additional neuroimaging studies employing an action fluency paradigm compared to generation of proper nouns, for instance, would also help elucidate brain regions critical to action word

PAGE 91

91 generation in particular. Contrasting patients with isolated focal or subcortical disease would also provide clarification the r elative role of these structures in action fluency performance. Finally, a more controlled study comparing patients with frontal lobe versus temporal lobe epilepsy would help to elucidate the contribution of these various structures to action fluency performance. To the best of our knowledge, this is the first study that has used a proper noun generation paradigm embedded within a fluency test as this construct has been studied spec ifically within a naming format to this point. The naming literature has offered preliminary support for this construct, possibly as mo re unique to anterior temporal functioning than tests employing naming of common objects. The results that we found are promising and provide converging evidence that this may indeed be a paradigm, both in the nami ng and fluency format that shows great pr omise with a variety of clinical populations. Additional research with this fluency and naming paradigm is absolutely warranted. First, further psychometric studies need to be undertaken to help establish the construct validity of th ese measures and sho w the convergent and discri minant validity with other tests in both clinically and normally -functioning populations. In order for the test to be used clinically, adequate normative studies need to establish performance in healthy controls on the fluency p aradigm. Development of a standardized naming paradigm could prove to be more difficult due to stimul us selection, which is something that was carefully considered in the present study. Naming stimuli have to be free of identifying features (i.e., uniforms for example), stratified across decade, and controlled for type and amount of fame (i.e., sports, news, politics, movies). As we foun d in the present study, familiarity with face stimuli will vary across person a nd must be accounted for This should be then be contrasted with naming ability in order to gain an

PAGE 92

92 accurate appreciation of true word retrieval deficits. Additional complications with developing a clinical famous faces naming task include the fact that it must be updated over time to account for famous persons now obsolete, or new famous persons in popular media. Nonetheless, the famous name paradigm, particularly the fluency format, should be studied in additional clinical epilepsy populations to document its utility with this sample. Prelimina ry studies could utilize post -ATL populations, patients with circumscribed lesions, to show the sensitivity and perhaps specificity to the anterior temporal lobe. The paradigm could then be advanced to pre -surgical populations to ascertain its ability to a id in the differential diagnosis of refractory seizure populations, the point of the present study. This paradigm could also offer significant value to other clinical populations apart from epilepsy. Similar to patients with left temporal lobe epilepsy, patients with Alzheimers d ementia evidence disease of temporal lobe structures and perform poorly on naming and semantic fluency paradigms. Name fluency may prove to be sensitive to degenerative disease of the temporal structures as well. Of more interest are patients with pre -Alzheimers, or mild cognitive impairment (MCI). Most of these patients evidence focal impairment in one domain, but later go on to progress to full blown dementia. As of late, the focus of research has turned to identifying these p atients in a pre -clinical state so that early intervention may be undertaken. As many of these patients repor t that their earliest problem i s difficulty retrieving names of people, this measure may provide a useful clinical tool in ident ifying early defici ts not tapped by other assessment measures Finally, o ur findings with regard to qualitative fluency performance were discordant than most of those presented in the literature, but may be explained by overall lack of fluency differences across groups. As this finding of related qualitative quantitative performance has been reported elsewhere, it is at least a likely possibility that these scores are dependent. Given

PAGE 93

93 that, the clinical utility of clusters and switches necessarily depends on how sensitive fluency measures are at detecting the presence or absence of pathology. Clustering and switching analyses are easy to compute and do not require significant additional work once familiarity with the scoring criteria has been established, making them relativ ely easy measures to consider clinically. However, additional work needs to be done to establish a normative basis for these measures before they are used in impaired populations At the present time, only one study has attempted to norm these measures (Tr oyer, 2000). Should normative data be established, these measures could be used clinically in conjunction with overall fluency score to help provide useful information about localization of function.

PAGE 94

94 APPENDIX A STANDARD NEUROPSYCHOLOGICAL TEST BATTER Y (SNB ) Construct Test Name and Reference Description of Test Measure Dependent Variables; Time to Administer Intellectual Functioning Wechsler Abbreviated Scale of Intelligence (WASI; Psychological Corporation, 1999) Block Design, Matrix Reasoning, Vocabulary, and Similarities subtests Verbal IQ, Performance IQ, Full Scale IQ; subtest T scores Time=20 30 Memory Functioning Rey Complex Figure Test and Recognition Trial (Myers & Myers, 2004) Measure of figural memory visuoconstruction that requires copy, immediate and delayed recall of a complex figure Scores for Immediate and Delayed Recall Time=15 California Verbal Learning Test 2nd Edition (Delis et al., 2000) Verbal list learning; assesses learning strategy, immediate and delayed recall, rec ognition, and interference Scores for Immediate and Delayed Free Recall Time=15 Wechsler Memory Scale R; Logical Memory I & III (Wechsler, 19 9 7) Measure of verbal memory for stories and figural memory for geometric designs. Scores for Immediate and D elayed Recall Time=15 Language Functioning Controlled Oral Word Association (COWA; Spreen & Benton, 1977) Verbal fluency for alphabet letter (i.e. F,A,S) Total correct exemplars Time=5 Semantic Fluency Animals (Tombaugh et al., 1999) Verbal fluenc y for a semantic category Total correct exemplars Time=5 Boston Naming Test II (Goodglass & Kaplan, 2000) Confrontation naming using large ink drawings Total Correct Time=10 Visuo perceptual / Visuo constructional Functioning WASI Block Design subt est (Wechsler, 1997) Visuoconstructional measure requiring construction with blocks Total score based on time limits Time=10 Frontal / Executive Skills -Attention, Psychomotor Speed, Abstract Thinking Wisconsin Card Sorting Test (Heaton, 1981) Measure of mental flexibility and problem solving. Number of categories achieved; Number of errors; trials to finish first category Time=15' Trail Making Test (Reitan, 1958) Measures visuomotor speed, set shifting Total time to complete; Number of errors Tot al time=5 WAIS III Digit Span (Wechsler, 1997) Memory for digit sequences; Requires attention span Total correct score Time=10 Sensory Perceptual and Motor Finger Tapping (Halstead, 1947; Reitan & Wolfson, 1993) Speeded fine motor movement for domina nt and non dominant hands Average number of taps across five trials Time=5 Grooved Pegboard Test (Klove, 1963) Speeded fine motor movement and dexterity Total time; # of drops Time=5 Mood and Affect Beck Depression Inventory II (Beck, Steer, & Brown, 1996 ) 21 item self evaluation questionnaire assessing elements of depression Total number of items endorsed Time=5

PAGE 95

95 APPENDIX B SAMPLE RESPONSES FROM FLUENCY DATA Name Fluency George Bush George Bush Sr. Bill Clinton Hillary Clinton John Kerr y John Edwards John McCain Barack Obama Beyonce Johnny Depp Kirstin Dunst Tom Hanks Meg Ryan Elizabeth Taylor Spencer Tracy Katherine Hepburn George Burns Gracie Allen George Harrison John Lennon Ringo Starr Paul McCartney Dakota Fanning Ricky Martin Actors/Actresses Presidents Politicians Temporally Grouped Musicians

PAGE 96

96 Action Fluency Run Jump Hop Skip Walk Step Jumping Jacks Jump Rope Hop Scotch Think Eat Drink Swallow Sip Sleep Lay Relax Rest Smile Blink Squint Think Kick Toss Hurl Throw Put Pedal Ride Reel Feet/legs Exercise Mouth Rest/Sleep Facial Gestures Hands

PAGE 97

97 REFERENCES Alexander, M. P., Stuss, D. T., Shallice, T., Picton, T. W., & Gillingham, S. (2005). Impaired concentration due to frontal lobe damage from two distinct lesion sites. Neurology, 65, 572579. Alvarez, J. A., & Emory, E. (2006). Executive f unction and the frontal lobes: A meta analytic review. Neuropsychol ogy Rev iews 16, 17 42. Annegers, J. F., Hauser, W. A., Coan, S. P., & Rocca, W. A. (1998). A population-based study of seizures after traumatic brain injuries. New Engl and J ournal of Med icine 338, 20 24. Anneger s, J. F., Hauser, W. A., Lee, J. R., & Rocca, W. A. (1995). Secular trends and birth cohort effects in unprovoked seizures: Rochester, Minnesota 19351984. Epilepsia, 36 575579. Baldo, J.V., Schwartz, S., Wilkins, D., & Dronkers, N.F. (2006). Role of fr ontal versus temporal cortex in verbal fluency as revealed by voxel based lesion symptom mapping. J ournal of the Int ernational Neuropsychol ogical Soc iety 12, 896900. Baldo, J. V., & Shimamura, A. P. (1998). Letter and category fluency in patients with frontal lobe lesions. Neuropsychology, 12, 259267. Barr, W. B., Goldberg, E., Wasserstein, J., & Novelly, R. A. (1990). Retrograde amnesia following unilateral temporal lobectomy. Neuropsychologia, 28, 243 255. Bechara, A., Damasio, H., & Damasio, A. R. (2000). Emotion, decision making and the orbitofrontal cortex. Cereb ral Cortex, 10 295307. Blenner, J.L., (1993). The discriminant capacity of the Stroop test in tumor neurosurgical patients and its relationship to the visual evoked potential measure. Personality and Individual Differences, 15, 99102. Bredart, S. (1993). Retrieval failures in face naming. Memory, 1 351366. Butler, M., Retzlaff, P.D., & Vanderploeg, R. (1991). Neuropsychological test usage. Professional Psychology: Research and Practice, 22, 510512. Butters, N., Granholm, E., Salmon, D. P., Grant, I., & Wolfe, J. (1987). Episodic and semantic memory: a comparison of amnesic and demented patients. J ournal of Clin ical and Exp erimental Neuropsychol ogy 9 479497. Cabeza R., & Nybe rg, L.(2000). Neural bases of learning and memory: functional neuroimaging evidence. Current Opinions in Neurology, 13, 415421.

PAGE 98

98 Cappa, S. F., Binetti, G., Pezzini, A., Padovani, A., Rozzini, L., & Trabucchi, M. (1998). Object and action naming in Al zheimer's disease and frontotemporal dementia. Neurology, 50, 351355. Caramazza, A., & Hillis, A. E. (1991). Lexical organization of nouns and verbs in the brain. Nature, 349, 788790. Chin, R. F., Neville, B. G., & Scott, R. C. (2005). Meningitis is a common cause of convulsive status epilepticus with fever. Arch ives of Disease in Child hood, 90 66 69. Corcoran, R., & Upton, D. (1993). A role for the hippocampus in card sorting? Cortex, 29 293304. Costello, A. L., & Warrington, E. K. (1989). Dynamic aphasia: the selective impairment of verbal planning. Cortex, 25 103114. Cotelli, M., Borroni, B., Manenti, R., Alberici, A., Calabria, M., Agosti, C., et al. (2006). Action and object naming in frontotemporal dementia, progressive supranuclear palsy, and corticobasal degeneration. Neuropsychology, 20, 558565. Damasio, A.R. (1995). On some functions of the human prefrontal cortex. Annals of the New York Academy of Science, 769, 241251. Damasio, H., Grabowski, T. J., Tranel, D., Ponto, L. L., Hichw a, R. D., & Damasio, A. R. (2001). Neural correlates of naming actions and of naming spatial relations. Neuroimage, 13(6 Pt 1), 10531064. Damasio, A. R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed neural systems. Proc eedings of the Nat ional Acad emies of Sci ence U S A, 90, 49574960. Damasio, H., Tranel, D., Grabowski, T., Adolphs, R., & Damasio, A. (2004). Neural systems behind word and concept retrieval. Cognition, 92, 179229. Daniele, A., Giustolisi, L., Silveri, M. C., Colosimo, C., & Gainotti, G. (1994). Evidence for a possible neuroanatomical basis for lexical processing of nouns and verbs. Neuropsychologia, 32, 13251341. Davies, K. G., Hermann, B. P., Dohan, F. C., Jr., & Wyler, A. R. (1996). Intractable epil epsy due to meningitis: results of surgery and pathological findings. Br itish J ournal of Neurosurg ery 10 567570. Delis, D.C., Kaplan, E., and Kramer, J.H. (2001). The Delis Kaplan Executive Function System. San Antonio: The Psychological Corporation.

PAGE 99

99 D e Nbrega E, Nieto A, Barroso J, Montn F. (2007). Differential impairment in semantic, phonemic, and action fluency performance in Friedreich's ataxia: possible evidence of prefrontal dysfunction. Journal of the International Neuropsychological Society 13, 94452. Demakis, G. J. (2004). Frontal lobe damage and tests of executive processing: a meta analysis of the category test, stroop test, and trail -making test. J ournal of Clin ical and Exp erimental Neuropsychol ogy 26 441450. Demakis, G. J., Mercu ry, M. G., Sweet, J. J., Rezak, M., Eller, T., & Vergenz, S. (2003). Qualitative analysis of verbal fluency before and after unilateral pallidotomy. Clin ical Neuropsychol ogy 1 7 322330. Diaz, M., Sailor, K., Cheung, D., & Kuslansky, G. (2004). Category size effects in semantic and letter fluency in Alzheimer's patients. Brain and Language 89 108114. Drane, D. L., Lee, G. P., Cech, H., Huthwaite, J. S., Ojemann, G. A., Ojemann, J. G., et al. (2006). Structured cueing on a semantic fluency task differe ntiates patients with temporal versus frontal lobe seizure onset. Epilepsy and Behavior, 9, 339344. Drane D L Ojemann G A Aylward E Ojemann J G Johnson L C. Silbergeld D L Miller J W & Tranel D. (2008). Category -specific naming and r ecognition deficits in temporal lobe epilepsy surgical patients. Neuropsychologia, 46, 124255. Exner, C., Boucsein, K., Lange, C., Winter, H., Weniger, G., Steinhoff, B. J., et al. (2002). Neuropsychological performance in frontal lobe epilepsy. Seizure 11 20 32. Forsgren, L., Beghi, E., Oun, A., & Sillanpaa, M. (2005). The epidemiology of epilepsy in Europe a systematic review. Eur opean J ournal of Neurol ogy 12 245253. Fossati, P., Guillaume le, B., Ergis, A. M., & Allilaire, J. F. (2003). Qualitative analysis of verbal fluency in depression. Psychiatry Res earch 117, 17 24. Frith, C. D., Friston, K. J., Herold, S., Silbersweig, D., Fletcher, P., Cahill, C., et al. (1995). Regional brain activity in chronic schizophrenic patients during the perf ormance of a verbal fluency task. Br itish J ournal of Psychiatry, 167, 343349. Frith, C. D., Friston, K. J., Liddle, P. F., & Frackowiak, R. S. (1991). A PET study of word finding. Neuropsychologia, 29, 11371148. Fukatsu, R., Fujii, T., Tsukiura, T., Ya madori, A., & Otsuki, T. (1999). Proper name anomia after left temporal lobectomy: a patient study. Neurology, 52, 10961099. Gleissner, U., & Elger, C. E. (2001). The hippocampal contribution to verbal fluency in patients with temporal lobe epilepsy. Cor tex, 37 55 63.

PAGE 100

100 Glosser G., & Donofrio, N. ( 2001). Differences between nouns and verbs after anterior temporal lobectomy. Neuropsychology, 15, 3947. Glosser, G., Salvucci, A. E., & Chiaravalloti, N. D. (2003). Naming and recognizing famous faces in te mporal lobe epilepsy. Neurology, 61, 81 86. Goldberg, E & Bougakov, D. (2005). Neuropsychologic assessment of frontal lobe dysfunction. Psychiatr ic Clin ics of North Am erica 28, 567580. Gorno Tempini, M. L., Price, C. J., Josephs, O., Vandenberghe, R., Cappa, S. F., Kapur, N., et al. (1998). The neural systems sustaining face and proper name processing. Brain, 121 (Pt 11) 21032118. Gourovitch, M. L., Kirkby, B. S., Goldberg, T. E., Weinberger, D. R., Gold, J. M., Esposito, G., et al. (2000). A compar ison of rCBF patterns during letter and semantic fluency. Neuropsychology, 14, 353360. Grabowski, T. J., Damasio, H., Tranel, D., Ponto, L. L., Hichwa, R. D., & Damasio, A. R. (2001). A role for left temporal pole in the retrieval of words for unique ent ities. Human Brain Mapping 13 199212. Harris, D.M., & Kay, J. (1995). I recognize your face but cant remember your name: Is it because names are unique? British Journal of Psychology, 86, 34558. Heaton R.K., Miller, S.W., Taylor M.J., & Grant, I. (2004) Revised Comprehensive Norms for an E xpanded Halstead Reitan B attery Psychological Assessment Resources, Odessa, FL Helms taedter, C. (2001). Behavioral aspects of frontal lobe e pilepsy. Epilepsy and Behav ior 2 384395. Helmstaedter, C., Kemper B., & Elger, C. E. (1996). Neuropsychological aspects of frontal lobe epilepsy. Neuropsychologia, 34, 399406. Henry, J. D., & Crawford, J. R. (2004 a ). A meta analytic review of verbal fluency performance in patients with traumatic brain injury. Neurops ychology, 18, 621628. Henry, J. D., & Crawford, J. R. (2004 b ). A meta analytic review of verbal fluency performance following focal cortical lesions. Neuropsychology, 18, 284 295. Hermann B. P Perrine, K Chelune G J Barr, W Loring D W Straus s E Trenerry M R & Westerveld M. (1999). Visual confrontation naming following left a nterior temporal lobectomy: a comparison of surgical approaches. Neuropsychology, 13 3 9.

PAGE 101

101 Hermann, B., & Seidenberg, M. (1995). Executive system dysfuncti on in temporal lobe epilepsy: effects of nociferous cortex versus hippocampal pathology. J ournal of Clin ical and Exp erimental Neuropsychol ogy 17 809819. Hermann, B., Seidenberg, M., Bell, B., Rutecki, P., Sheth, R.D., Wendt, G., O'Leary, D., & Magnotta V. (2003) Extratemporal quantitative MR volumetrics and neuropsychological status in temporal lobe epilepsy. Journal of the International Neuropsychological Society, 9, 353362. Hermann, B. P., Seidenberg, M., Dohan, F. C., Jr., Wyler, A. R., Haltiner, A., Bobholz, J., et al. (1995). Reports by patients and their families of memory change after left anterior temporal lobectomy: relationship to degree of hippocampal sclerosis. Neurosurgery, 36, 3944; discussion 4435. Hermann, B. P., Seidenberg, M., Hal tiner, A., & Wyler, A. R. (1995). Relationship of age at onset, chronologic age, and adequacy of preoperative performance to verbal memory change after anterior temporal lobectomy. Epilepsia, 36 137145. Hermann, B. P., Seidenberg, M., Schoenfeld, J., & Davies, K. (1997). Neuropsychological characteristics of the syndrome of mesial temporal lobe epilepsy. Arch ives of Neurol ogy 54, 369376. Hermann, B. P., Wyler, A. R., & Richey, E. T. (1988). Wisconsin Card Sorting Test performance in patients with complex partial seizures of temporal lobe origin. J ournal of Clin ical and Exp erimental Neuropsychol ogy 10 467476. Ho A.K., Sahakia, B.J., Robbins, T.W., Barker, R.A., Rosser, A.E., & Hodges, J.R. (2002). Verbal fluency in Huntingtons disease: A longitudi nal analysis on phonemic and semantic clustering and switching. Neuropsychologia, 40, 12771284. Janowsky, J. S., Shimamura, A. P., & Squire, L. R. (1989). Source memory impairment in patients with frontal lobe lesions. Neuropsychologia, 27, 10431056. Joanette, Y., & Goulet, P. (1986). Criterion-specific reduction of verbal fluency in right brain damaged right handers. Neuropsychologia, 24, 875879. Jokeit, H., & Ebner, A. (2002). Effects of chronic epilepsy on intellectual functions. Prog ress in Brain Res earch 135, 455463. Lamar M. & Resnick, S.M. (2004). Aging and prefrontal functions: dissociating orbitofrontal and dorsolateral abilities. Neurobiology of Aging, 25, 553 338. Lezak, M.D., Howieson, D.B. & Loring, D.W. (2004). Neuropsychological As sessment (4th ed.). New York: Oxford University Press.

PAGE 102

102 Loring, D.W., Marino, S., & Meador, K.J. (2007). Neuropsychological and behavioral effects of antiepilepsy drugs. Neuropsychological Review, 17, 413425. Lu, L. H., Crosson, B., Nadeau, S. E., Heilman, K. M., Gonzalez Rothi, L. J., Raymer, A., et al. (2002). Category-specific naming deficits for objects and actions: semantic attribute and grammatical role hypotheses. Neuropsychologia, 40, 16081621. Luria, A.R. (1966). Higher cortical function s in man. New York: Basic Books. Martin, A., & Fedio, P. (1983). Word production and comprehension in Alzheimer's disease: the breakdown of semantic knowledge. Brain and Language 19 124141. Martin, R. C., Loring, D. W., Meador, K. J., & Lee, G. P. (1 990). The effects of lateralized temporal lobe dysfunction on formal and semantic word fluency. Neuropsychologia, 28, 823829. Martin R. C. Sawrie S M Gilliam F G Palmer C. A Faught E Morawetz, R. B & Kuzniecky R I. (2000). Wisconsin Card S orting perform ance in patients with temporal l obe epilepsy: clinical and neuroanatomical correlates. Epilepsia, 41 162616 32. Martins, I. P., & Farrajota, L. (2007). Proper and common names: a double dissociation. Neuropsychologia, 45, 17441756. McCar thy, R., & Warrington, E. K. (1985). Category specificity in an agrammatic patient: the relative impairment of verb retrieval and comprehension. Neuropsychologia, 23, 709727. McDonald, C. R., Bauer, R. M., Filoteo, J. V., Grande, L., Roper, S. N., & Gilm ore, R. (2006). Episodic memory in patients with focal frontal lobe lesions. Cortex, 42 10801092. McDonald, C. R., Delis, D. C., Norman, M. A., Tecoma, E. S., & Iragui, V. J. (2005). Discriminating patients with frontal lobe epilepsy and temporal lobe e pilepsy: utility of a multilevel design fluency test. Neuropsychology, 19, 806813. McDonald, C. R., Delis, D. C., Norman, M. A., Tecoma, E. S., & Iragui -Madozi, V. I. (2005). Is impairment in set -shifting specific to frontal lobe dysfunction? Evidence fr om patients with frontal -lobe or temporal lobe epilepsy. J ournal of the Int ernational Neuropsychol ogical Soc iety 11 477481. McDonald, C. R., Delis, D. C., Norman, M. A., Wetter, S. R., Tecoma, E. S., & Iragui, V. J. (2005). Response inhibition and set shifting in patients with frontal lobe epilepsy or temporal lobe epilepsy. Epilepsy and Behav ior 7 438446. McKenna, P., & Warrington, E.K. (1980). Testing for nominal dysphasia. Journal of Neurology, Neur osurgery, and Psychiatry, 43, 781788.

PAGE 103

103 McMilla n, A.B., Hermann, B.P., Johnson, S.C., Hansen, R.R., Seidenberg, M., & Meyerand, M.E. (2004). Voxel -based morphometry of unilateral temporal lobe epilepsy reveals abnormalities in cerebral white matter. Neuroimage, 23, 167174. Miceli, G., Silveri, M. C., Villa, G., & Caramazza, A. (1984). On the basis for the agrammatic's difficulty in producing main verbs. Cortex, 20 207220. Milders, M. (2000). Naming famous faces and buildings. Cortex, 36 138145. Miller, E. (1984). Verbal fluency as a function of a measure of verbal intelligence and in relation to different types of cerebral pathology. Br itish J ournal of Clin ical Psychol ogy 23 ( Pt 1) 53 57. Monsch, A. U., Bondi, M. W., Butters, N., Salmon, D. P., Katzman, R., & Thal, L. J. (1992). Comparisons of verbal fluency tasks in the detection of dementia of the Alzheimer type. Arch ives of Neurol ogy 49 12531258. Mummery, C. J., Patterson, K., Hodges, J. R., & Wise, R. J. (1996). Generating 'tiger' as an animal name or a word beginning with T: differenc es in brain activation. Proc eedings: Biological Sci ence, 263, 989 995. N'Kaoua, B., Lespinet, V., Barsse, A., Rougier, A., & Claverie, B. (2001). Exploration of hemispheric specialization and lexico -semantic processing in unilateral temporal lobe epilepsy with verbal fluency tasks. Neuropsychologia, 39, 635 642. Nichols, M.E., Meador, K.J., & Loring, D.W. (1993). Neuropsychological effects of antiepileptic drugs: a current perspective. Clinical Neuropharmacology, 16, 471484 Obler L.K., & Albert, M. L. ( 1979). The Action Naming Test (Experimental ed.) Boston: VA Medical Center. Oldfield, R.C. (1971). The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia, 9, 97113. Ostberg, P., Fernaeusm S.E., Hellstrmm K., Bogdanov (2005). Impaired verb fluency: a sign of mild cognitive impairment. Brain and Language, 95, 273279. Owen, A., Downes, J., Sahakian, B., Polkey, C., & Robbins, T. (1990). Planning and spatial working memory following frontal lo be lesions in man. Neuropsychologia, 28, 10211034. Oyegbile, T.O., Bhattacharya, A., Seidenberg, M., & Hermann, B.P. (2006). Quantitative MRI biomarkers of cognitive morbidity in temporal lobe epilepsy. Epilepsia, 47, 143152.

PAGE 104

104 Paradowski, B., & Zagraje k, M. M. (2005). Epilepsy in middle aged and elderly people: a three year observation. Epileptic Disord ers, 7 91 95. Parks, R.W., Loewenstein, D.A., Dodrill, K.L., Barker, W.W., Yoshii, F., Chang, J.Y., Emran, A., Apicella, A., Sheramata, W.A., & Duara, R. (1988). Cerebral metabolic effects of a verbal fluency test: A PET scan study Journal of Clinical and Experimental Neuropsychology, 10, 565575. Paulesu, E., Goldacre, B., Scifo, P., Cappa, S. F., Gilardi, M. C., Castiglioni, I., et al. (1997). Funct ional heterogeneity of left inferior frontal cortex as revealed by fMRI. Neuroreport, 8 20112017. Pendleton, M. G., Heaton, R. K., Lehman, R. A., & Hulihan, D. (1982). Diagnostic utility of the Thurstone Word Fluency Test in neuropsychological evaluations. J ournal of Clin ical Neuropsychol ogy 4 307317. Pran, P., Cardebat, D., Cherubini, A., Piras, F., Luccichenti, G., Peppe, A., Caltagirone, C., Rascol, O., Dmonet, J.F., & Sabatini, U. (2009). Object naming and action-verb generation in Parkinson's disease: A fMRI study. Cortex [Epub ahead of print] Available online March 14, 2009. Perani, D., Cappa, S. F., Schnur, T., Tettamanti, M., Collina, S., Rosa, M. M., et al. (1999). The neural correlates of verb and noun processing. A PET study. Brain, 122 ( Pt 12) 23372344. Perret, E. (1974). The left frontal lobe of man and the suppression of habitual responses in verbal categorical behaviour. Neuropsychologia, 12, 323 330. Perucca, E. (2005). An introduction to antiepileptic drugs. Epilepsia, 46 Sup pl 4 31 37. Petersen S.E., Fox, P.T., Posner, M.I., Mintum, M.A., & Raichle, M.A. ( 1989). Positron emission tomographic studies of the processing of single words. Journal of Cognitiv e Neuroscience, 1 153170. Petersen, S. E., Fox, P. T., Snyder, A. Z. & Raichle, M. E. (1990) Activation of extrastriate and frontal cortical areas by visual words and word like stimuli. Science, 249, 1041 1044. Petrides M. & Milner, B. ( 1982) Deficits on subject -ordered tasks after frontal and temporal l obe lesi ons in man. Neuropsychologia, 20, 24962. Piatt, A. L., Fields, J. A., Paolo, A. M., Koller, W. C., & Troster, A. I. (1999). Lexical, semantic, and action verbal fluency in Parkinson's disease with and without dementia. J ournal of Clin ical and Exp eriment al Neuropsychol ogy 21 435443.

PAGE 105

105 Piatt, A. L., Fields, J. A., Paolo, A. M., & Troster, A. I. (1999). Action (verb naming) fluency as an executive function measure: convergent and divergent evidence of validity. Neuropsychologia, 37, 14991503. Raichle, M. E., Fiez, J. A., Videen, T. O., MacLeod, A. M., Pardo,J. V., Fox, P. T. & Petersen, S. E. (1994) Practice related changes in human brain functional anatomy during non -motor learning. Cerebral Cortex 4, 8 26. Ramier, A. M., & Hecaen, H. (1970). [Respect ive roles of frontal lesions and lesion lateralization in "verbal fluency" deficiencies]. Rev Neurol (Paris), 123(1), 17 22. Randolph, C., Braun, A.R., Goldberg, T.E., & Chase, T.N. (1993). Semantic fluency in Alzheimers, Parkinsons, and Huntingtons Di sease: Dissociation of Storage and Retrieval Failures. Neuropsychology, 7, 8288. Rapp B., & Caramazza, A. ( 1998). A case of selective difficulty in writing verbs. Neurocase, 4 127139. Rende, B., Ramsberger, G., & Miyake, A. (2002). Commonalities and differences in the working memory components underlying letter and category fluency tasks: a dual task investigation. Neuropsychology, 16, 309321. Reverberi, C., Laiacona, M., & Capitani, E. (2006). Qualitative features of semantic fluency performance in mesial and lateral frontal patients. Neuropsychologia, 44, 469 478. Risberg, J., & Grafman, J. (2006). The frontal lobes: Development, function, and pathology Cambridge, UK: Cambridge University Press. Robert P.H., Lafont, V., Medecin, I., Berthet, L., Thauby, S., Baudu, C., & Darcourt, G. ( 1998). Clustering and switching in verbal fluency tasks: Comparison between schizophrenics and healthy adults. Journal of the International N europsychological Society, 4, 539546. Rosser, A., & Hodges, J. R. (1 994). Initial letter and semantic category fluency in Alzheimer's disease, Huntington's disease, and progressive supranuclear palsy. J ournal of Neurol ogy, Neurosurg ery, and Psychiatry, 57 13891394. Ruff, R. M., Light, R. H., Parker, S. B., & Levin, H. S (1997). The psychological construct of word fluency. Brain and Language 57 394405. Seidenberg, M., Griffith, R., Sabsevitz, D., Moran, M., Haltiner, A., Bell, B., et al. (2002). Recognition and identification of famous faces in patients with unilater al temporal lobe epilepsy. Neuropsychologia, 40, 446456. Semenza, C., & Zettin, M. (1989). Evidence from aphasia for the role of proper names as pure referring expressions. Nature, 342, 678679.

PAGE 106

106 Shapiro, K. A., Moo, L. R., & Caramazza, A. (2006). Cortic al signatures of noun and verb production. Proc eedings of the Nat ional Acad emies of Science U S A, 103, 16441649. Silveri, M. C., Salvigni, B. L., Cappa, A., Della Vedova, C., & Puopolo, M. (2003). Impairment of verb processing in frontal variant -frontot emporal dementia: a dysexecutive symptom. Dement ia and Geriatr ic Cognitive Disord ers, 16 296 300. Sperling, M. R. (2004). The consequences of uncontrolled epilepsy. CNS Spectr ums, 9 (2), 98 101. Spreen, O. & Strauss, E. (1998). A Compendium of Neuropsychological Tests. New York: Oxford University Press. Stuss, D. T., & Alexander, M. P. (2000). Executive functions and the frontal lobes: a conceptual view. Psychol ogical Res earch 63 (3 4), 289298. Stuss, D. T., Alexander, M. P., Hamer, L., Palumbo, C., De mpster, R., Binns, M., et al. (1998). The effects of focal anterior and posterior brain lesions on verbal fluency. J ournal of the Int ernational Neuropsychol ogical Soc iety 4 265278. Stuss, D. T., Floden, D., Alexander, M. P., Levine, B., & Katz, D. (200 1). Stroop performance in focal lesion patients: dissociation of processes and frontal lobe lesion location. Neuropsychologia, 39, 771786. Stuss, D. T., & Levine, B. (2002). Adult clinical neuropsychology: lessons from studies of the frontal lobes. Annua l Rev iews of Psychol ogy 53 401433. Stuss, D. T., Toth, J. P., Franchi, D., Alexander, M. P., Tipper, S., & Craik, F. I. (1999). Dissociation of attentional processes in patients with focal frontal and posterior lesions. Neuropsychologia, 37, 10051027. Thaiss, L. & Petrides, M. (2003), Source versus content memory in patients with a unilateral frontal cortex or a temporal lobe excision. Brain, 126, 11121126. ThompsonSchill, S. L., Swick, D., Farah, M. J., D'Esposito, M., Kan, I. P., & Knight, R. T. (1998). Verb generation in patients with focal frontal lesions: a neuropsychological test of neuroimaging findings. Proc eedings of the Nat ional Acad emies of Sci ence U S A, 95, 1585515860. Tranel, D. (1992). Neuropsychological assessment. Psychiatr ic Cl in ics of North Am erica 15 28399. Tranel, D. (2006). Impaired naming of unique landmarks is associated with left temporal polar damage. Neuropsychology, 20, 1 10.

PAGE 107

107 Tranel, D., Damasio, H., & Damasio, A.R. (1997). A neural basis for the retrieval of conceptual knowledge. Neuropsychologia, 35, 13191327. Tranel, D., Damasio, H., Eichhorn, G.R., Grabowski, T., Ponto, L.L., & Hichwa, R.D. (2003). Neural correlates of naming animals from their characteristic sounds. Neuropsychologia, 41, 847854. Tran el, D., Grabowski, T. J., Lyon, J., & Damasio, H. (2005). Naming the same entities from visual or from auditory stimulation engages similar regions of left inferotemporal cortices. J ournal of Cognitive Neurosci ence, 17 12931305. Trenerry, M., & Jack, C. R. Jr. (1994). Wisconsin Card Sorting Test performance before and after temporal lobectomy. Journal of Epilepsy, 7, 313317. Troster, A. I., Salmon, D. P., McCullough, D., & Butters, N. (1989). A comparison of the category fluency deficits associated w ith Alzheimer's and Huntington's disease. Brain and Language 37 500 513. Troster, A. I., Warmflash, V., Osorio, I., Paolo, A. M., Alexander, L. J., & Barr, W. B. (1995). The roles of semantic networks and search efficiency in verbal fluency performance in intractable temporal lobe epilepsy. Epilepsy Res earch 21 19 26. Troyer, A.K. (2000). Normative data for clustering and switching on verbal fluency tasks. Journal of Clinical and Experimental Neuropsychology, 22, 370378. Troyer, A. K., & Moscovitch M. (2006). Cognitive processes of verbal fluency tasks. In Poreh, Amir M. (ed), The Quantified Process Approach to Neuropsychological Assessment (pp. 143160). New York: Taylor & Francis. Troyer, A. K., Moscovitch, M., & Winocur, G. (1997). Clustering and switching as two components of verbal fluency: evidence from younger and older healthy adults. Neuropsychology, 11, 138146. Troyer, A. K., Moscovitch, M., Winocur, G., Alexander, M. P., & Stuss, D. (1998 a ). Clustering and switching on verbal fluency: the effects of focal frontal and temporal lobe lesions. Neuropsychologia, 36, 499504. Troyer, A. K., Moscovitch, M., Winocur, G., Leach, L., & Freedman, M. (1998 a ). Clustering and switching on verbal fluency tests in Alzheimer's and Parkinson's disease J ournal of the Int ernational Neuropsychol ogical Soc iety 4 137143. Tsukiura, T., Fujii, T., Fukatsu, R., Otsuki, T., Okuda, J., Umetsu, A., et al. (2002). Neural basis of the retrieval of people's names: evidence from brain -damaged patients and fMRI. J ournal of Cognitive Neurosci ence, 14 922 937.

PAGE 108

108 Tyler L.K., Bright, P., Flecther, P., & Stamatakis, E.A. ( 2004). Neural processing of nouns and verbs: the role of inflectional morphology. Neuropsychologia, 42, 512523. Vendrell, P., Junque, C., Pujol, J., Jurado, M. A., Molet, J., & Grafman, J. (1995). The role of prefrontal regions in the Stroop task. Neuropsychologia, 33, 341352. Vilkki, J., & Holst, P. (1994). Speed and flexibility on word fluency tasks after focal brain lesions. Neuropsychologia, 32, 12571262. Wakamoto, H., Hayashi, M., Nagao, H., Morimoto, T., & Kida, K. (2004). Clinical investigation of genetic contributions to childhood-onset epilepsies and epileptic syndromes. Brain & Dev elopment 26 184 189. Warrington, E. K. (2000). Homop hone meaning generation: a new test of verbal switching for the detection of frontal lobe dysfunction. J ournal of the Int ernational Neuropsychol ogical Soc iety 6 643648. Warrington, E. K., & Shallice, T. (1984). Category specific semantic impairment s. Brain, 107 (Pt 3) 829 854. White -Devine T., Grossman, M., & Robinson, K.M. ( 1996). Verb confrontation naming and word -picture matching in Alzheimers disease. Neuropsychology, 10, 495503. Woods, S. P., Carey, C. L., Troster, A. I., & Grant, I. (2 005). Action (verb) generation in HIV 1 infection. Neuropsychologia, 43, 11441151.

PAGE 109

109 BIOGRAPHICAL SKETCH Bonnie C. Sachs received a bachelors degree in p sych ology from Virginia Tech and a masters degree in behavioral n euroscience from American Universit y She worked for two years as a research assistant at the National Institutes of Health prior to entering the doctoral program in the Department of Clinical and Health Psychology at the University of Florida. Bonnie earned her masters degree in clinical psychology from the University of Florida in 2005, and completed her clinical internship at the Department of Rehabilitation Medicine at Emory University during the 20082009 academic year Bonnie received her doctoral degree in c linical psychology (neuropsychology track) from the University of Florida in 2009. Currently she is employed as a postdoctoral fellow at the Mayo Clinic Her main research interests include the neuropsychology of epilepsy, patterns of cognitive impairment in dementia, neurorehabil itation, and functional neuroimaging.