<%BANNER%>

Influences of Mediated Priming on Tip-of-the-Tongue Incidence and Resolution

Permanent Link: http://ufdc.ufl.edu/UFE0021558/00001

Material Information

Title: Influences of Mediated Priming on Tip-of-the-Tongue Incidence and Resolution
Physical Description: 1 online resource (66 p.)
Language: english
Creator: Merrill, Lisa A
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Previous research indicates that tip-of-the-tongue (TOT) states, characterized by the temporary inability to retrieve a known word, can be reduced as well as resolved via phonological priming (i.e., presentation of prime words phonologically related to the TOT target word). The current study investigated the influences of mediated phonological priming, in which the prime and target were indirectly related, on the incidence and resolution of TOT states. Participants answered general knowledge questions designed to induce a TOT state and named pictures in which one was either a prime or unrelated picture. The prime was a near-synonymous picture that corresponded to a dominant (e.g., motorcycle) and secondary (e.g., bike) name, where the dominant name referred to the name typically produced upon seeing the picture. The secondary name was phonologically related to the TOT word by sharing its first syllable (e.g., biopsy). In Experiment 1, participants named a near-synonymous picture before answering a general knowledge question, and the results showed that producing the dominant name of prime pictures increased target retrieval and decreased TOT states for subsequently presented targets, relative to naming unrelated pictures. In Experiment 2, participants named a near-synonymous picture after answering the general knowledge questions, and for targets that were in TOT states, naming the prime pictures' dominant name increased resolution of those TOT states compared to unrelated pictures. Because production of the dominant name led to decreased TOT incidence and increased TOT resolution than did an unrelated name, these results demonstrate that the phonologically-related secondary name (without being produced) primed the target for retrieval. In addition to providing another method of priming TOT states for future research, exposure to the TOT word's first-syllable phonology via mediated priming may be a process involved in the spontaneous resolution of TOT states in everyday life.
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 Lisa A Merrill.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Abrams, Lise.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021558:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021558/00001

Material Information

Title: Influences of Mediated Priming on Tip-of-the-Tongue Incidence and Resolution
Physical Description: 1 online resource (66 p.)
Language: english
Creator: Merrill, Lisa A
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Previous research indicates that tip-of-the-tongue (TOT) states, characterized by the temporary inability to retrieve a known word, can be reduced as well as resolved via phonological priming (i.e., presentation of prime words phonologically related to the TOT target word). The current study investigated the influences of mediated phonological priming, in which the prime and target were indirectly related, on the incidence and resolution of TOT states. Participants answered general knowledge questions designed to induce a TOT state and named pictures in which one was either a prime or unrelated picture. The prime was a near-synonymous picture that corresponded to a dominant (e.g., motorcycle) and secondary (e.g., bike) name, where the dominant name referred to the name typically produced upon seeing the picture. The secondary name was phonologically related to the TOT word by sharing its first syllable (e.g., biopsy). In Experiment 1, participants named a near-synonymous picture before answering a general knowledge question, and the results showed that producing the dominant name of prime pictures increased target retrieval and decreased TOT states for subsequently presented targets, relative to naming unrelated pictures. In Experiment 2, participants named a near-synonymous picture after answering the general knowledge questions, and for targets that were in TOT states, naming the prime pictures' dominant name increased resolution of those TOT states compared to unrelated pictures. Because production of the dominant name led to decreased TOT incidence and increased TOT resolution than did an unrelated name, these results demonstrate that the phonologically-related secondary name (without being produced) primed the target for retrieval. In addition to providing another method of priming TOT states for future research, exposure to the TOT word's first-syllable phonology via mediated priming may be a process involved in the spontaneous resolution of TOT states in everyday life.
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 Lisa A Merrill.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Abrams, Lise.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021558:00001


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 E20101211_AAAACE INGEST_TIME 2010-12-12T03:49:54Z PACKAGE UFE0021558_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 6978 DFID F20101211_AACBON ORIGIN DEPOSITOR PATH merrill_l_Page_23thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
b98b1e31286f15cdfd6cb9f1f68721eb
SHA-1
74b70faf757b6f41512430ba8a70c2be65ae184c
54949 F20101211_AACBJP merrill_l_Page_20.jp2
6bdb396d666dac39c88eddea0ff1899b
7db0e5468fbcf118eb76e5eb8bacd446df71664f
77086 F20101211_AACBES merrill_l_Page_41.jpg
57ff5bfba9e5bd3b7a5825a3e47ae935
31a8e42b7b9e5bea66b93e0b3b62ec00c0a636d5
25271604 F20101211_AACAXV merrill_l_Page_43.tif
6bf2c9ae2c99bee900f44648b8c0a356
73736e16bdb4d997598f7213029b3db020e182c1
6719 F20101211_AACBOO merrill_l_Page_26thm.jpg
5814d5286fdff7a35c8ebc5bef93c569
1cf35760ae4b23714908ff7f9f1f0d697fbb5f64
301591 F20101211_AACBJQ merrill_l_Page_32.jp2
e65ef8bb4ac63515349f611204da19da
be9340f229136cc343a71d725e87b7b9f22f0bb7
30353 F20101211_AACBET merrill_l_Page_35.jpg
19e72a7923e7ae734b613ab180a2c61f
db888f62972fa7f462716f3a912d5385c0a99cf7
2642 F20101211_AACAXW merrill_l_Page_32thm.jpg
f26f0a3c7a452e3a04406f0d29553db9
2a74d2ef7763cd9a817138cb582a0cfb58de1329
26496 F20101211_AACBOP merrill_l_Page_27.QC.jpg
31863899eb3f332e82c8496c493d17dd
516bf30adcbfed26657051a11d13f6a111ab3124
231071 F20101211_AACBJR merrill_l_Page_33.jp2
fbbb96df9dd2ebf5c98dffdc223bfaf3
4aff6ee98f23439a4f54c719a67b9a2e34ad14c1
F20101211_AACBEU merrill_l_Page_01.tif
c6bfa3e6bcf04d53a2faaf93f810e92c
b4412cfd6d75f1da4e36018eb42148b73ed43bde
1051979 F20101211_AACAXX merrill_l_Page_22.jp2
d9dbbbcb09aeff97c3a9dd663600da40
60a7b8fe8c19357d92f531cab8afffc3271abeb6
6976 F20101211_AACBOQ merrill_l_Page_29thm.jpg
c1ceec4e7f50873631b804dce165d706
cd4ea52d2f0d8a7d32006e962737c37308049384
231781 F20101211_AACBJS merrill_l_Page_34.jp2
739b0faeb5f5be05d51b109a4a3f84ed
c58549e54d02f0f2aeaa319f9f399df8523fd631
2067 F20101211_AACBEV merrill_l_Page_26.txt
ffbf7f58547fb2beff88d28461750215
9c8169ca14cfdca49d079f4f819e5d9b53d5cc44
488972 F20101211_AACAXY merrill_l_Page_07.jp2
b7c8eefe5fe1a2a13927a1978dcbaa1b
ab803c971730bb7d5e1a9a911c291bc90a9c555a
7060 F20101211_AACBOR merrill_l_Page_30thm.jpg
35b503342c6a75f51181b3b7e93b606b
2cc69c8b15fe560fa61e4833aa7aa0ac2919d00f
486767 F20101211_AACBJT merrill_l_Page_36.jp2
c796ac11d247579510ec7d0ce3474796
c2a6100cf7d1962979301535a136a08c0ccd1ffa
F20101211_AACBEW merrill_l_Page_27.tif
304da6acfb40ef5f1f1d3ec0910f61c8
e98e1d91da2545cc7393837f29920587efb2950b
1051975 F20101211_AACAXZ merrill_l_Page_41.jp2
559c0ca9c283f8f69d4d5264e18ebd85
23f71a00d65a7e25dcc1868bab2fe1afef82dafe
7466 F20101211_AACBOS merrill_l_Page_31thm.jpg
6d1841cd4ac0f1916c1ab746e37d9c07
d2b18326d6c7c755946bd2fdbb48037d2f535a57
1051952 F20101211_AACBJU merrill_l_Page_37.jp2
f2b37127298b0631614b26f1c4c89692
e5339a0cdfd92c3459195b3dee75ad46dd3e83dc
F20101211_AACBEX merrill_l_Page_10.tif
6f3e559c054ebde47a93fa8af2c1023c
98c731fad49291ab22c3a10128a2b5a02af7f69f
8734 F20101211_AACBOT merrill_l_Page_32.QC.jpg
9ef63ce3ffccc35946f06395111d2b48
dcbaf3fb27d3358f17ade859d3e32409c58b1c40
1051986 F20101211_AACBJV merrill_l_Page_43.jp2
c4ebe7527c97deaf38f99840a9b8111d
b40080d8214a84c1161dded7cd976b02a038170a
7252 F20101211_AACBCA merrill_l_Page_27thm.jpg
2790a13dddefc17bd4cf1bac6cf86c4e
194a5a67e004d132ff61b338a3ba93736c6145cc
26690 F20101211_AACBEY merrill_l_Page_62.QC.jpg
52e4c22b868beb03ad478ffe245b389e
5240e9e1fbfc05de7ddf4ddfafe81eddea2d64f3
89997 F20101211_AACBCB merrill_l_Page_58.jpg
c55ee8f0f692496dbae2ea654a8c982b
c521ea61b5bcf919e7dc503bf76a5ebf69d00f01
497134 F20101211_AACBEZ merrill_l_Page_04.jp2
15d4bda045e7cddc3913202f346f5a89
46a1f56a6deaaf09f408a496810a84917e655b96
1051966 F20101211_AACBJW merrill_l_Page_45.jp2
a39c69bf36c60ec40273022aceac86a3
b465fa26759e3d4ad6dd8e686df11d0bac19b9e7
2841 F20101211_AACBOU merrill_l_Page_35thm.jpg
6ed4549e65876dac407b74a2dfa1f1ae
abffbe4a6912eecbed1ed6eb17edaa5b0bfc6c59
85555 F20101211_AACBCC merrill_l_Page_16.jpg
fe03c68d21621b6905e6fc953016ec23
bf1223610fb9fea4474c2b075d5b88a9ff7834ab
211005 F20101211_AACBJX merrill_l_Page_50.jp2
ee69071db1a6f59b99333b32ef58e451
48c33efc7dada1c4bd924ea613bb3064adaa5c80
6693 F20101211_AACBOV merrill_l_Page_38thm.jpg
2e863558d444d8474ad93d8d9c5e5d2c
5a15306872b5ded68771f0816d5eec2527f64ac8
57724 F20101211_AACBCD merrill_l_Page_61.jpg
ad8bd200cfcdc03f2c2f6f6326a0d445
64e71298e03cf267796275977a45844366f58398
35163 F20101211_AACBHA merrill_l_Page_61.pro
3c500083ec8c5e33f8aaca2a6ca7979c
6137a010ae6eb875851ff57454d8c587e60adb2f
617260 F20101211_AACBJY merrill_l_Page_52.jp2
6058b27add763b237ed8d36cddfdba7c
e2460133ca1adda3d898b99cdcd264a746130f8f
7232 F20101211_AACBOW merrill_l_Page_40thm.jpg
04e20254dd5047eb5ac0465a72053356
da9d4136a438485e4bf9929bb6cff1a2eb0f389d
1239 F20101211_AACBCE merrill_l_Page_48.txt
fb84a13f6855776addad0839941bec09
804abdc5f93e477cd2cf2125370d18c594f5481e
287887 F20101211_AACBHB merrill_l_Page_51.jp2
94bd26159997f63a455c3492f0b72d17
f2d3d07f7c39573f5c3890f1a59431426f2dcd6a
1051889 F20101211_AACBJZ merrill_l_Page_56.jp2
a25038ad32888592c511bd01258cb6b1
bc57921f3cecc6c1605b7d67669e7c76c7a8b8bf
24188 F20101211_AACBOX merrill_l_Page_41.QC.jpg
ffb62dbb2323df770335d32f35e6b9f1
270177d9df13020b1afdd29cee1526e475fd1be2
12805 F20101211_AACBCF merrill_l_Page_20.QC.jpg
bb275e41780b419fe7852c1b25bfacaa
31076896585dd090214a1645cd2a592efdbc0b48
1051968 F20101211_AACBHC merrill_l_Page_54.jp2
3f5ab84e7a6961e84f145ef9d49fc01e
60b7605ee4317af5b6064934fc95b286292dca16
2789 F20101211_AACBOY merrill_l_Page_51thm.jpg
5f3035d4118c2d97cabbf28184965c10
5fb899bb9707e14ba767845e9ae86614037a96f2
56574 F20101211_AACBCG merrill_l_Page_55.pro
0969225300f27f48de374302c9e571e4
f345c61bf7fabce50aafbfb51b62425bee76444c
898 F20101211_AACBMA merrill_l_Page_04.txt
49b80d86b594718c51f77f22a94a945a
4ae31c1ea8777cb554f81dfa03b63c16bf7b2966
26060 F20101211_AACBHD merrill_l_Page_40.QC.jpg
5d409dfd779d1491772a9fd95a71f1c2
529ae28036c98bd3c34c963d1dc09171152488ee
8391 F20101211_AACBOZ merrill_l_Page_53.QC.jpg
921f8703794bc263f1bb45b48783c7ca
ecaa6b5fc15db34567bd89078edf58e238954ea7
371 F20101211_AACBCH merrill_l_Page_49.txt
b860dbe7f53fb7fc9d7c0c8c40eafc72
d5f6dcb779782d6ae191adebee704d5cb5e48b58
1667 F20101211_AACBMB merrill_l_Page_05.txt
2212d8ec6af123038b538484cd999f63
8f73812b87df0a9d3fc2dab93a5682af2f0425c4
31117 F20101211_AACBHE merrill_l_Page_48.pro
5b66c95d6d105ff5e8e55d858e9b4e5d
edae099c12e7054a891fa23861345ec273fdea93
1330 F20101211_AACBMC merrill_l_Page_06.txt
b3472a462ec2e7af73f2546adf342633
a1aedbcd189e4d1a40fb32836c47a11cca9d8c81
1311 F20101211_AACBHF merrill_l_Page_03thm.jpg
744565f5e1f4b5a6dd1e1c139c36da21
3382b6463c2c353da40d5656c8345dacced6e362
232315 F20101211_AACBCI merrill_l_Page_66.jp2
cba36c380579e95e18436517262457be
4e1c333cf85294a7c39da74a00999ccad7cef8d5
535 F20101211_AACBMD merrill_l_Page_07.txt
3e7f04f6810de44de6777fe618ecdd30
afa8582107dbf1b45eb3f6431abddd6e75bd9218
7258 F20101211_AACBHG merrill_l_Page_46thm.jpg
2b78883aa4451c8c00175b5111b63dbe
074b2ed6ee0eae6384995daf6fededd8a0ad5924
2130 F20101211_AACBME merrill_l_Page_10.txt
529304e1082f738372f175132b273340
909fb0918b9f244665f904dfd3ecf0a4c9660cae
6916 F20101211_AACBHH merrill_l_Page_25thm.jpg
a66620b8f399f9c291f6d2c04c8ebbf5
54986138f7094944478d46754c43de55ac19af6e
7043 F20101211_AACBCJ merrill_l_Page_43thm.jpg
1d1955ef03c898cf876df6e7b428d913
82b770865c1ac5e7337ac6f7ffac74ac5099772d
2177 F20101211_AACBMF merrill_l_Page_11.txt
297655e6ad8451a3999e6913cdb56ef7
746a5ca6b4cf4ace9fc598e792b98f51762ab448
25230 F20101211_AACBHI merrill_l_Page_26.QC.jpg
fb528dd7485afd689c6f82071155d1e7
a98755864757d00c22f9f830ca3494276335b579
18125 F20101211_AACBCK merrill_l_Page_03.jp2
d7b236974854ffd90be5b73e55adfdc9
a488b2b65e96d5034b015f492f3fe34418ac12c9
2142 F20101211_AACBMG merrill_l_Page_14.txt
0766ffb0a61429165202cfc9c9d5427b
3d8cbe210fd5c6690f21d4ff5d59184ec061a9e1
25433 F20101211_AACBHJ merrill_l_Page_07.jpg
151cd7bf71c49c08bdee9b18aae91f18
397e3bf642303a3d831f7eb13f2d7c28e511e024
236998 F20101211_AACBCL merrill_l_Page_53.jp2
c36e12b711d2a2d5403b9908e8186e73
f7ec37139a590b8c3cd292c39cc6c7335e70c35e
2151 F20101211_AACBMH merrill_l_Page_16.txt
558f9120fc82667aea72b8ac003ef0df
ef7e3a47903d9fa4f04f9371f814ba67d489afc0
86563 F20101211_AACBHK merrill_l_Page_22.jpg
aeae558b47847fba614987c64550f4e8
35f51b01655bd674d0a7a44ce6148f908d6304db
7737 F20101211_AACBCM merrill_l_Page_07.QC.jpg
4daaa0af42578a587e7fb8d3bd04cafe
8bec8fd537be1df042687c7f50b3a11339203185
2132 F20101211_AACBMI merrill_l_Page_17.txt
87ae3de0142af635ec1e89e6b1dbb42d
fbb548199ea22ee58083bc898eaa666f880a299f
1051978 F20101211_AACBHL merrill_l_Page_12.jp2
59c5e615934fefd618da135336600a6d
3a1e7310e1fb52d4a9a0c17c689033e482cfe931
85661 F20101211_AACBCN merrill_l_Page_11.jpg
fde189ed94f32246a664c20081b3c689
b974243b5e99efc5f8cdb1061d9d8a7084bb0f8e
1895 F20101211_AACBMJ merrill_l_Page_19.txt
7a33d9cbf9254585fdb4ea9a1f882e2e
60a393a08d94bad42ec127d127966122f4923587
16612 F20101211_AACBHM merrill_l_Page_61.QC.jpg
ba8497695fcd77f74c1abf6c03972301
92fffe1c47631cef65409e1510e8b49f65bd708d
12356 F20101211_AACBCO merrill_l_Page_51.pro
ee89fffed052ed2e3961136007d6dc10
ecb20d67ea314a4574b53804b629dd81512af797
1892 F20101211_AACBMK merrill_l_Page_21.txt
79223541f7ca4a62d7cbd4a9d85470eb
126e7a92a29938aa84c39a185630a401cb714313
F20101211_AACBCP merrill_l_Page_06.tif
fc7b6762bbb60563d5048c217b3cef4b
4fb61c30b6a7cc2818a8512896757c2634159aba
2141 F20101211_AACBML merrill_l_Page_22.txt
415cea12cf8b5ca443874a3bc98ad9b4
be38af1345917dca8bf34df9332c0bd2603f6f76
1051969 F20101211_AACBHN merrill_l_Page_06.jp2
23ac27ae656fc858ba44a6363ccf596b
7e988cf4e38c582cc004080e934c2010d2028cb8
88138 F20101211_AACBCQ merrill_l_Page_57.jpg
2cb54226fd670f207eac60101eda094c
4122b898da5b053527f16065f9fcfb55a7ac05da
2032 F20101211_AACBMM merrill_l_Page_25.txt
77b5f834bdd47e3dfdfb24b2a947560a
88cb15ef287035988869240f50fb5c48e711163f
2169 F20101211_AACBHO merrill_l_Page_13.txt
05d8c9d37d3b870d4cc450f80cb0116c
f2b35c0b72798f383eea3dc75b2aed53e789441a
1889 F20101211_AACBCR merrill_l_Page_38.txt
819d4f79f9c56875eb468e3cdbffd949
f29b7304dde79b41e468399c943aac071e9ed8c9
1998 F20101211_AACBMN merrill_l_Page_27.txt
4dac670cad078f44e55e6a1ff2e58fa4
ff6c168ce6311d8d2860d288d95f6ea9f40adf01
7301 F20101211_AACBHP merrill_l_Page_28thm.jpg
4e6c3a1f61756c0aec31085a9b9e6306
31a14c11250bda39045a6471658b91aaeb865e17
53824 F20101211_AACBCS merrill_l_Page_31.pro
95498db58a9a8cc8c2c4681822eee993
0cd183b480ece4ce39ea1c5ed6c1fc40151cefaf
2098 F20101211_AACBMO merrill_l_Page_28.txt
e0bd745c052594547247bb18d9798318
11407ba1677f4b17ee39a6d1c2605dc837e45d56
440 F20101211_AACBHQ merrill_l_Page_01.txt
2bb03f4fcf739d43cf7663eb8b7fec04
46e7f61c08ab2417c518e70d86bf5318c83a9ac9
25575 F20101211_AACBCT merrill_l_Page_25.QC.jpg
9b490385065f85f29cfabfb06a6a3fc8
8e9984351f7e0a5096a6d15e8fca4a05cfd80d22
1903 F20101211_AACBMP merrill_l_Page_29.txt
1942a3883820e87046973e814ea69fdf
75be903883ba189d26bf1a67800ff984edd119f5
7683 F20101211_AACBHR merrill_l_Page_34.QC.jpg
c2e66d1b100859b3850a59b32447850c
782917dfcf199ee0b29c6624bf3a9bdd05ddbdee
F20101211_AACBCU merrill_l_Page_64.tif
0ca7bfc0a3192f9bc7e043a40bae19a4
6085eef6014d1b32fbd5299fe5cc71c0c0c9f2d0
587 F20101211_AACBMQ merrill_l_Page_35.txt
37402c141f86782bd50d67e7f4fd95ad
c858a1eb0cb5b66460650563d93d10fb38771d1e
784 F20101211_AACBHS merrill_l_Page_20.txt
0d8c483f1e9df848b58d69757d990c41
9531ac664fd7dbd6acaea1cabfc2786428883b07
7551 F20101211_AACBCV merrill_l_Page_47thm.jpg
fc9e593531099008c7733efa33558bfc
5eecb0cb3799deec3176e93112642b5c23144285
2084 F20101211_AACBMR merrill_l_Page_40.txt
ff737d6f2c24ffb13c1b267d1cce390b
7e2745371866296e88df2eb004929a82302a5334
82377 F20101211_AACBHT merrill_l_Page_23.jpg
e2a5b7251b1006faeb92f2b0a1429a13
3505a4601689826b4f7dc8db0e740e0b83faf3e0
58165 F20101211_AACBCW merrill_l_Page_58.pro
0b9864647b0c8cfcb0c8341560d0a3c6
7ac79dbc782dcba031c4061913184f94957fc685
57840 F20101211_AACBHU merrill_l_Page_12.pro
6bfef1fa39461efd4b333f03c46aacbb
cee6f8b234e2cb67b15b898e2017c0044e5b89b2
446 F20101211_AACBCX merrill_l_Page_34.txt
fbe0f98c5e78278b8d1343299587621a
11385ccaa5be013ece506cb9425f2b8a4ad25b3f
28605 F20101211_AACBHV merrill_l_Page_58.QC.jpg
e45d4c61a5451dec0346d55ac348b0bc
d3bfc178f65c7794913d111e8fa6d214679d740a
87853 F20101211_AACBAA merrill_l_Page_54.jpg
d3bc55adaff06c74305675532596de6a
3f4b9d82b8842a251baf85bcddd76bf38f7dca36
F20101211_AACBCY merrill_l_Page_16.tif
2cde36f7389a21f350a772ab4c250d1e
6dbdfe37dfe163874cc6330763f175c238f63e12
2081 F20101211_AACBMS merrill_l_Page_44.txt
f674e286043fdfb2f73d7ead7c6e82be
4121e11cc7c69295240bca1c440d22ce371ceb9b
2127 F20101211_AACBMT merrill_l_Page_45.txt
c6a793dbf4e930ed235a79f5387253a4
40b86d07d05b349f73f242606b6302ca8bf2d533
2631 F20101211_AACBHW merrill_l_Page_34thm.jpg
b18e8dd6e1227d990a13142881c7b2d3
5a90f8e5fd0eb63fac7cec8806b8f8df6d74a2b4
1051983 F20101211_AACAYA merrill_l_Page_11.jp2
5a7dd6b60aa704471925a980f2cb92e1
f06ca03fdc50a332f3277980c2f41ca6584044fa
F20101211_AACBAB merrill_l_Page_19.tif
36d28f676a9ad1187d56e471a4747c81
d94f5701a5c8055f5eaf05bc6ce0f2e0487943da
2076 F20101211_AACBCZ merrill_l_Page_59.txt
1de8b07c26103444e1a683cde83f0382
0be78e367e31bc20e8aec0879ec0b70ca3cd1320
2086 F20101211_AACBMU merrill_l_Page_46.txt
b63e297f23e1f5e7a5174af98c8fc0ca
17fdabcb54b0780dbcf4de85f0ebaf1bd080efdd
51219 F20101211_AACBHX merrill_l_Page_26.pro
9129e25a0fff1f2033083baff1c9fc09
24c8400bd847e1b5e2167372a21d569b77667c48
622 F20101211_AACAYB merrill_l_Page_03.pro
627f249b9bfc3acac0e2a713bc09bb4c
534512fca6f2e20d842379da60e958a35b053c5b
F20101211_AACBAC merrill_l_Page_41.tif
cfab798adfa1761d710b1247599d76ba
ad3261d4c42fffcaa7bcce8c8a6943900392cf99
523 F20101211_AACBMV merrill_l_Page_53.txt
07a433b534f4c782dac07373deaf7222
910f9d2c2c377e7a1d44a662ff8773b02b3808be
7433 F20101211_AACBFA merrill_l_Page_11thm.jpg
4ba8521c7fbd7ab349894559a9e3ca09
bed9c1ace0dd475499e87d69ac0fdfe5042acaf0
1051971 F20101211_AACBHY merrill_l_Page_29.jp2
97de69a1ad2c99659dcbdc96754a7348
2d9a7155944e7fee6627aa5bc4527d99b2dd82e6
14824 F20101211_AACAYC merrill_l_Page_05.QC.jpg
7e591fe1874f32b04c974d35adf3dff7
5ff8d295587f2a19564c45b389c721954d548ca9
F20101211_AACBAD merrill_l_Page_29.tif
1046778d0c504fe142654a0e010e7899
a1dcb960bb59bbb8f0e8891020d5317168bf3cc9
2240 F20101211_AACBMW merrill_l_Page_54.txt
94edd55ac1c053a382d3183795608494
2d29fd7c52a7053e8eee197deda7e12f706fefe9
F20101211_AACBFB merrill_l_Page_59.jp2
77ca982f12055d76aa794fe6fcbcbd41
03dff26690dc164ec2386fb90e62e3a1c716ed30
2224 F20101211_AACBHZ merrill_l_Page_55.txt
45e1c76ce9ba32e628320faa851d9582
e26aa27ca6c5566967b3e6eba6818d0e22d63456
7345 F20101211_AACAYD merrill_l_Page_10thm.jpg
84102434389e789c44e70741aa94f2f8
7f5728d26cd9d0fb6b7eedefa7ad53ab189b6adf
2117 F20101211_AACBAE merrill_l_Page_31.txt
ecc41af7b8ea15b76d6477f436c3cce3
d09905e0fb1c020ece34eabbcf290f3f3bf8976f
2190 F20101211_AACBMX merrill_l_Page_57.txt
046fbc80ad2d0ef3801b1f300cc26d8d
e1d3dcd00fa6f67de452e015fffad0d56440a694
85056 F20101211_AACBFC merrill_l_Page_28.jpg
f9eea0450b9b7720559a56fb075b3f94
cac4b0437a19d5f99a88b749a708d920fb77183a
1051980 F20101211_AACAYE merrill_l_Page_18.jp2
caf87b362d406ab3f9d43b59dbd0a6b3
939150bf2df039fc62e9f5d5053454673d643b3b
449 F20101211_AACBAF merrill_l_Page_33.txt
f42223ef45108cbbc15d0570f86465d1
7ffac3bb9cf1aeb91079e8c0742654dccc0fe644
2274 F20101211_AACBMY merrill_l_Page_58.txt
05cb27be710e83102b3e57c6043200e8
cbdbbc45198e4ad0751107435f1e4d13efd646a7
53489 F20101211_AACBFD merrill_l_Page_28.pro
175cf88836bd04736bf19ef48e609a61
202ec4dac56b795e1fb168ff94728ab8c6704863
84632 F20101211_AACAYF merrill_l_Page_59.jpg
c39c9ef646000cf0c5f688c1bffc3295
3e3809532880e534fbd6ecd52d70251f2bb56b1f
490395 F20101211_AACBKA merrill_l_Page_63.jp2
0f75e077c29c64ddd5350870800c3890
b2420446fd4eb1025153e9962078b790bbdcf972
1428 F20101211_AACBMZ merrill_l_Page_61.txt
536770d68a57dd84d93bd85ae2252dc5
ca059eaa25e1d9b19928431fb2764652d2f77cb1
1037556 F20101211_AACBFE merrill_l_Page_38.jp2
73f9bd19546494d56d1a7a1376cc9da2
152d8dde13ccfdcd6b9997d2ea3e7d7427918de9
5379 F20101211_AACAYG merrill_l_Page_60thm.jpg
98709a3d98a36945b993cb317e31b66b
6c9a37d52c8fc209fd5df9fcb0a3661da397ca3b
101 F20101211_AACBAG merrill_l_Page_02.txt
ac5b80d2e3141627b6a06890543395e6
b5798f1678298a3fed889ff37ab4d27ac2acd258
1051954 F20101211_AACBKB merrill_l_Page_65.jp2
db1b000181388869af20ca78a5d0b84b
0c1dc1c92fbab9083c61aee40191f7d1d55c33f4
76398 F20101211_AACBFF merrill_l_Page_08.jpg
3c5fc3bf800e9e6238a8fb5064fec8bf
77b12a5fe750eba20b16f6b19e4b0eb74f79a826
26959 F20101211_AACAYH merrill_l_Page_42.QC.jpg
800b163c59ac5c1f090c35de065b6774
58cb891a4b5cceb00e76570231191cc692b2898c
54158 F20101211_AACBAH merrill_l_Page_17.pro
efa5657897132ebea611891779b7e792
dd62b65574f92725fc66cd60cabfd7b6f4d47c12
F20101211_AACBKC merrill_l_Page_11.tif
ca3778d0c55e853e2cdd5720bbabc20a
0fc8f2c377c458a3fe79508936e2a00db9f8ca6f
27901 F20101211_AACBPA merrill_l_Page_56.QC.jpg
87beca8242e17aee03c4043f3cc03c63
219c2207c0ba009b676b5bd18c896ca86a08b6b5
1051950 F20101211_AACAYI merrill_l_Page_64.jp2
3674822562d1f02c3645cc7578b29ef7
c34bcecf733a52d6300e2d85ecac96d8101302b0
25068 F20101211_AACBAI merrill_l_Page_53.jpg
64435fd9357de16eae2ebec7675f08d4
48b37d79781c5e1e2fe51e87f8265a0d4c6dc734
F20101211_AACBKD merrill_l_Page_12.tif
351663cc31bee5114d433e2fa0ff6bbc
0653eaff7fad7d65f7e3ea99a5fa3b8c71870975
1051985 F20101211_AACBFG merrill_l_Page_46.jp2
c8c6d058b9281751f1651fd9b4bc1d50
877b173ad8aee3e3d50bab9bf4bc01e61959a4a2
18926 F20101211_AACBPB merrill_l_Page_60.QC.jpg
aa16c07110c72c2cf28f287777638f75
9693cddbbe6163f36ffc908edabe5254eee4544d
13685 F20101211_AACAYJ merrill_l_Page_35.pro
fb89476b359e36999b8506cb6482d78d
8a35f9af03964513a963f4ace3c236c1b135c870
F20101211_AACBAJ merrill_l_Page_31.jp2
c75c13facc3ca5a5fe9b1b6fc8e2d77b
b1ac32ce782ac42fcb1d3a98990c9e1ad4a4f368
F20101211_AACBKE merrill_l_Page_18.tif
dbd3b888f5635fdc89dcfa18959e5f23
09910efd147594af320123bce605640caf670096
884 F20101211_AACBFH merrill_l_Page_63.txt
ba711ccbb580b0b3325717ae75933862
c2d8ef9aa71b1d9cafab3c84092bd789383b78bb
4880 F20101211_AACBPC merrill_l_Page_61thm.jpg
0281c0ad4f09624c78ad3175b37f88cb
a772cbc738d0e162575b1a749e748a92c1c2a952
2138 F20101211_AACAYK merrill_l_Page_18.txt
1121cc2703a8c358ee247b15f64439d1
13d6d474806560a8e9c8ee01321042efc703f09a
F20101211_AACBAK merrill_l_Page_14.jp2
f0e23d2bb341db57cbdd65efa7955bb0
62c56a09229e8b3244d81d83bb1f81514f5bb072
1053954 F20101211_AACBKF merrill_l_Page_20.tif
be1268d46aba6cf064d99301a3b09f6e
919a7d2ca44849faaf5da7c33cb7d2db7597cd01
66329 F20101211_AACBFI merrill_l_Page_21.jpg
e3379f4c876f83522c998fc031588ed3
27e38389cf13a3f1a5400f0eb1694671b2d2f378
12018 F20101211_AACBPD merrill_l_Page_63.QC.jpg
afc7720f7096c0169cacedc92ba2c017
bb1faac62d8863d8826ffa43d597d152c59a906a
F20101211_AACAYL merrill_l_Page_13.tif
f0dea6ff852b1b506b403b5e06f6c2f1
93c53a6a170bba9a32d45376db2993b8522661e0
F20101211_AACBAL merrill_l_Page_07.tif
2b184175fc7501b88a2e3fda65f2bb38
6626bb9c6f0a6c8eb1261f9ec02d2a177967913f
F20101211_AACBKG merrill_l_Page_26.tif
d510fd5beefa70349ab31dfe8792852a
b4f4b11264fe306a0ef8250055591b8b5b1edda4
1051959 F20101211_AACBFJ merrill_l_Page_57.jp2
7e98ad867df7dc5121e2306cca0a73a9
308d4e3dfcc43fbe5a8eb8351d80e4de2b18c795
25914 F20101211_AACBPE merrill_l_Page_64.QC.jpg
7833eeffdb62ff8eeb43957ea892a5d3
a7eb19badfd5e1f962ee8076eee13940d99e9638
48256 F20101211_AACAYM merrill_l_Page_41.pro
926764eef8806a795ac36aaddf575e5e
7d6aa7916e6aed1777b72c106deb082559b7dd88
7531 F20101211_AACBAM merrill_l_Page_58thm.jpg
27cb05ab5baed9f98ed4172ab5fd6212
f76d63b9ac9035350daa242223852c7fe782dea0
F20101211_AACBKH merrill_l_Page_31.tif
5b7a0b33d58e8b3186b9304eb8fe0ced
ee4b2a773f7971a281186faeacf40ec5a2005ab1
1033572 F20101211_AACBFK merrill_l_Page_08.jp2
5d71b11637a8c1de370b62e7cf90f6a7
01f78b01d4a16e30421eb3b43acc2b6d30ba22dc
7391 F20101211_AACBPF merrill_l_Page_66.QC.jpg
0aedd81857b16e3a52f7c01dcc42775a
1f1c66ddcec0ed15d2147f8360770b87263b26bc
832109 F20101211_AACAYN merrill_l_Page_60.jp2
e37c33496306d5650289ebec539b6402
1a216399df85cda807f6ade89e308d4e9483d943
83401 F20101211_AACBAN merrill_l_Page_44.jpg
b869afb40a5ace1c7552f8688c2b9443
560ad2e7838a79ec9526e505a5072a5493e450b7
F20101211_AACBKI merrill_l_Page_33.tif
5745b2dcd32fc4e90ea61652b3d114de
5737a17171208e46013c42cf5cfa76b51fc023d9
2778 F20101211_AACAYO merrill_l_Page_65.txt
5d574c3af7ab44c576ee2a84c67900d5
0e8fd0597aba7b4f6fb77f3d904d960f65b40f31
1913 F20101211_AACBAO merrill_l_Page_41.txt
21b2be15dd2eb1c9b175484e2f8c74cb
e7ee3eb52514545703e4e8eff520f0bdb66821de
F20101211_AACBKJ merrill_l_Page_34.tif
e5d807e0b80dfdaa4645340fa4e81874
a36335413572e84426750d3d09e799bcdac11fc4
7439 F20101211_AACBFL merrill_l_Page_42thm.jpg
84129e2e4ae4406d70dfd4299c1ace43
8ec4133d6a3c3bdf8affc88ed53e8b2b8c8e86fa
26687 F20101211_AACAYP merrill_l_Page_59.QC.jpg
cacb1e5335984cb3e0d115c557d75ec1
e87001ef0de4604b78cc4bc6e775a76cca67f8ef
F20101211_AACBAP merrill_l_Page_39.jp2
366cd28b9d01a4c99af3954af3ff71d5
3bbfa7b807f82526700d5217ea3868fce4b60456
F20101211_AACBKK merrill_l_Page_36.tif
0af5a0c4d1221158ad83f208bc513b6e
6d0984e5f8c17098257b128f29bc53b85ebd1598
55070 F20101211_AACBFM merrill_l_Page_47.pro
970a217b1cf5b2a1aae987d19c549847
0ecb80aba34c8796c0a6b46cb8ea37e0baafd380
1051967 F20101211_AACAYQ merrill_l_Page_58.jp2
29f97854680e4f3711762d8d77f3344f
8691f0b7fc9ececd59a6b57761755b5e9ba71604
710934 F20101211_AACBAQ merrill_l_Page_48.jp2
b08dc45a7a8422ae8b689a6e83302294
ab96f37c3d0f95d3f960fd0042a1cee52ec2a82f
F20101211_AACBKL merrill_l_Page_40.tif
02398a813875843d4ddefc7d4882d0d8
c16507b13f6db1ca275e62923a8c117f1709605d
9124 F20101211_AACBFN merrill_l_Page_51.QC.jpg
37572a1a6fa35c7c76f4d971d0bd0cf7
a73e1ad0c0d852ac7b4ff5a94da503ff09e89e9a
364 F20101211_AACAYR merrill_l_Page_50.txt
e792766ac3288685f3b407b664ca70a4
2fa88cb704794e157301573553d178a24aaa4375
F20101211_AACBAR merrill_l_Page_22.tif
6d9e8498dfb517a6780104ccb1de5104
d117d0d145121ee638c19556c13ebf99709adaca
F20101211_AACBKM merrill_l_Page_42.tif
f37f0f0d5b9da47b684e8320ad82990b
cd25fe8540903039d959c36fac04ddfe5627e687
7481 F20101211_AACBFO merrill_l_Page_57thm.jpg
9b9513be21fdce5ce42b1809f23855f7
d1ddc56bb673e9ffb1107d835fe697884f8f6d7f
9248 F20101211_AACAYS merrill_l_Page_35.QC.jpg
9df503b6c74c714fb20705c91646f6c3
37237be236e00874e39079ccb01eed24dbb01128
24503 F20101211_AACBAS merrill_l_Page_34.jpg
f1bed233d965567561a88ed1e9929e26
8643231bf9286e98e273c8b448a5908838fde025
F20101211_AACBKN merrill_l_Page_44.tif
5a1c231a56916ff7275578ef6853a439
10fa7c595a3f9630519eae7c2c0645de00a7546a
1051981 F20101211_AACBFP merrill_l_Page_27.jp2
a262469f559137055912ea63179cbc68
cc7cf807a768ded6bf0d380d07a200502feafe30
24891 F20101211_AACAYT merrill_l_Page_43.QC.jpg
f71aaf6647a63949f21fe3d0458db56f
5b750a8943fe4ac99a9dce256246a0844bf826fc
50841 F20101211_AACBAT merrill_l_Page_24.pro
d6e91b911c2a8c9e00d32d5145effa19
1afaa96a316a00482711679d3865f118dcc91a21
F20101211_AACBKO merrill_l_Page_45.tif
9008b441012acca32be93414a7f12df0
4c482e4d7c2ee2224c64cf827b943f02a57ea7a3
651 F20101211_AACBFQ merrill_l_Page_09.txt
ee1acfc76ebb50d2c292f692a7320d88
10a19ce18be620dfafc0bcd1dd0f6a1b63087e17
80653 F20101211_AACAYU merrill_l_Page_25.jpg
ed871dba24e3234824387fcf714a36aa
e66d7f3c13027eb7cf4b3ef6458b8b8e8696d665
F20101211_AACBAU merrill_l_Page_28.tif
ce5da55722e364653b45c53010d78ccb
7bfae24b56450989cec1a753beb4ee5f85215340
F20101211_AACBKP merrill_l_Page_51.tif
a8d078ca6ceef6280ea83d1730a21e36
cc15260ae06c6657cdaedf7dd654f9e66fac415e
F20101211_AACBFR merrill_l_Page_15.tif
c8f8d80f1ae5b15be09d71f5c1d55f5f
4d026eb16ed004362559575008c4bf761a34ca91
12702 F20101211_AACAYV merrill_l_Page_04.QC.jpg
9aed36201d54a414bc19d97bcdb5ca04
9dd419d3dc372be7a4af41084bdc9bed2b725695
F20101211_AACBAV merrill_l_Page_58.tif
0dba76a930149a4c7c947c1cb1e891e2
80d3578f4b7afa7b3a667f9e656129e89107af68
53488 F20101211_AACBFS merrill_l_Page_18.pro
b7faf3527d6212a4187aa67a36ed11a0
1ef84bb432ae708f063646b13c617445922d6911
84779 F20101211_AACAYW merrill_l_Page_31.jpg
2e955f580853ee6285426ccd2051b73b
184917f308130b19d08051fa899621d3b48345f4
1051960 F20101211_AACBAW merrill_l_Page_10.jp2
f15822a14c308b31cb944092796ebf74
b0135f92f328b8b768167839cb7f3dc1e91bf99c
F20101211_AACBKQ merrill_l_Page_53.tif
bd7da3d206cfcd45002cfc7a3b2f5a0a
2a16aa9bc71faba6b34010387dbf8eba04f19b63
2002 F20101211_AACBFT merrill_l_Page_08.txt
ba3ed8a6a5390afbf4e1362bcdcf6f7b
e0d32fb12133ae925a2f1f50c48932884453bae4
85564 F20101211_AACAYX merrill_l_Page_42.jpg
0680c2a75312f66e99bd2569508922bd
74a075ca7da9661d25082c11caa4e9ce48198b25
83876 F20101211_AACBAX merrill_l_Page_40.jpg
fe40645e07d955ec42eab177dc5818df
a26765d5cf42b4bab769a390cf9020b77e362863
F20101211_AACBKR merrill_l_Page_55.tif
a9ddc5770f8a41bc2e5a351fe2dbc71e
50de1223f272318cda9d47f1852c4958b8be1262
26254 F20101211_AACBFU merrill_l_Page_44.QC.jpg
ff4905509d658b079004245127a0b196
520576c771e22b3290c4d52b9d00f5ad4bb49ff9
F20101211_AACAYY merrill_l_Page_65.tif
28dc1cd35173b7202f364756186bd474
dde13b7a951985b7991f0a91e7f3f43c2a558ca4
27268 F20101211_AACBAY merrill_l_Page_31.QC.jpg
86ae963dcc2c43dcb99364416baf7de7
9517bcc253e21a3aa3534327551f41790a364a9b
F20101211_AACBKS merrill_l_Page_56.tif
2492665d7450f0f51dc81ff97c50020d
c980f6785582843fc073416e3c1938e94d9b11c7
F20101211_AACBFV merrill_l_Page_02.tif
cb9d41f1e69bf2cf7af1fa7ac0720462
20dcb2505a021d2ace2d02ac888b43ea1269fdd5
F20101211_AACAYZ merrill_l_Page_21.tif
decded78af3ad77b613628522cad21f9
84f7c8fe8837afcba4b7d0340a6b651b43c3455e
F20101211_AACBAZ merrill_l_Page_08.tif
58899420a154cc1c30f9385c75d35f0c
9c95caa73e40708cccaa4026507d33582b32eae8
F20101211_AACBKT merrill_l_Page_57.tif
0702555a8b919fab06f7d5b6c321b06f
80fa13d3c5306ddf85f3fb6d481d8691c9c60a0d
F20101211_AACBFW merrill_l_Page_37.tif
72bbd86816a84ef855cf1940b3c51f4f
9dae2d7ea741fd4de47b9c4823797f3c2cdafa71
F20101211_AACBKU merrill_l_Page_59.tif
0e8f943a6bed2991e821a8ea36308189
1c61990371183be594f27070ccb9c8ad8a5279d1
7555 F20101211_AACBFX merrill_l_Page_65thm.jpg
19766ac810041a360c57cb18771dc6c4
e946b9a6091dae78c2eaea3870a22be51b5e1bb8
F20101211_AACBKV merrill_l_Page_61.tif
f667248535d0926c27bac795edbda8ed
799507c192b02daada31f0596486831404277b9d
90545 F20101211_AACBDA merrill_l_Page_12.jpg
8b42a7a40ad36ef68ff54f686f1828c7
eff981929c2a09dcd5b330462a8cb2be551bd152
F20101211_AACBFY merrill_l_Page_52.tif
04995799b90666600e20ca3fb0f599e5
20907dde97d045c36daf32251483dbfa3db090bb
F20101211_AACBKW merrill_l_Page_62.tif
8b47492f78c5a5eebc27efcac5cc9c21
2aa4ae09050cab8fe5f69c0c3bc4de36f68bff22
17024 F20101211_AACBDB merrill_l_Page_48.QC.jpg
850593c9ae6f05bbc2d4125d732943e6
d05c901a1a6afa8074d28d786aacd67ae54ba143
2158 F20101211_AACBFZ merrill_l_Page_47.txt
241f96aadba1e4d05164c4a1509b1747
df4f1f875b384b28e5a02583a4171b76572763ac
F20101211_AACBKX merrill_l_Page_63.tif
2d58ac553525d093a20fdc458e1cbb88
16066db1587735790fe428e0605f9f172572d4b4
12656 F20101211_AACBDC merrill_l_Page_07.pro
489a52e1d79562c51ce02f777b4477b0
a404de1e61d953f2b08217f2870a1c384fba07cc
F20101211_AACBIA merrill_l_Page_04.tif
2049cab435cea364215560b5c3fe6090
8df78628d235895eb8f97ff810705f7f6418095a
856 F20101211_AACBKY merrill_l_Page_02.pro
2510db992774347e4c2833f1e57de157
c5e05274d1ab478f8da6257fcee6443fca749fb9
F20101211_AACBDD merrill_l_Page_24.tif
15e5b96d95cc75fb221bbf3fc7bbf55b
0cc7a8d8e8c135e52deb0274dd5e82dcc17bf4c5
26666 F20101211_AACBIB merrill_l_Page_23.QC.jpg
3499492236ae95340f211281d343437b
81c7cdde4eff398db52c4c1ed98817cea4c14ee9
35992 F20101211_AACBKZ merrill_l_Page_05.pro
40fd0ff6a32ba7d87270e0e988e60f79
8978dac7fad32d6f326bb7f811fe0c9ca934e1d3
26711 F20101211_AACBDE merrill_l_Page_28.QC.jpg
d8f52434bcbeb5756ec68caffca628ee
b697c3aca2af411b79a8d91fe4299c9d24ca2800
6893 F20101211_AACBDF merrill_l_Page_41thm.jpg
c11d8019f7e4912735b466341339b5e3
30da89cde2084c72573f0b56f1d13be0a25909c0
F20101211_AACBIC merrill_l_Page_33.pro
a1b5dcc7deb9cc9ab9853186d4227707
1e028fba6044af3f978f02b635a269db24b6cd06
2426 F20101211_AACBNA merrill_l_Page_62.txt
90ad9960a049fda1a93d505f88a1b36a
82eea00368e177a2120f24b950ef3728fa1e348c
3808 F20101211_AACBDG merrill_l_Page_04thm.jpg
9997308e04ab076d37b25cc6c24784b5
034eb32f98bf1bea7e8723d1051ae4aee7f972b5
80360 F20101211_AACBID merrill_l_Page_24.jpg
ccf4e8db24dd3399f1f89c91bd1f274b
80bc78b45356c1552dd930273b5b7ed32fbf012e
2341 F20101211_AACBNB merrill_l_Page_64.txt
975eb7705bfc7dd7fa8e0b8d043ca6e6
18b85d2dd50e5d244aa2178ae2e55f509d8a2424
7541 F20101211_AACBDH merrill_l_Page_45thm.jpg
c76dbefa2d1217c694fc9108618a7c48
80b9a3900fd47a6d7f978e9353adc383d4396b79
78609 F20101211_AACBIE UFE0021558_00001.mets FULL
8cc24d5159c90674716110a09bbd6d2c
1a4d5a02101c06550634c1b35f8cf928894e565c
423 F20101211_AACBNC merrill_l_Page_66.txt
123ec3428c9d593a0aca174a23e12f20
362d2cabb3c04c4d3d1652d63c8255c59de8b64a
845 F20101211_AACBDI merrill_l_Page_36.txt
32473b412bbb076b4a05648bc3527aa3
f9aff2818a02848f9c2d0155aa1ed5be390f6a7f
7335 F20101211_AACBND merrill_l_Page_59thm.jpg
91c5431822414473deb53b2ab805f65a
1e23c33878c8b56027820e487dcc4e277cbf5854
21329 F20101211_AACBNE merrill_l_Page_21.QC.jpg
a689e2742b7157c64aadc3a963185cd5
53ba57632802f6f515a273b7e97d5ea4e710e3e0
9626 F20101211_AACBDJ merrill_l_Page_66.pro
78e938d01cdf22588870b6db96e6cabb
daae1ef08764b1250e97fe9a18be0a54038061f3
9930 F20101211_AACBIH merrill_l_Page_02.jpg
31b9203be479e4f14b4bf783873b9be9
5418e8d7b6912b6f26d15a7c7e02387482eac616
14122 F20101211_AACBNF merrill_l_Page_06.QC.jpg
4ed2da4ce8bf6d1248ecd6138e3b8994
1763c64ff8c1968fd89a9d51680adba311593793
F20101211_AACBDK merrill_l_Page_49.tif
ec1a91d1b303322765cbd07024c44f23
7a0fda391ecd6dc5624476a920ed267228fb53d5
9549 F20101211_AACBII merrill_l_Page_03.jpg
ae604bb8e836ea588dcf7b44dbd90e1c
9e4a84bd8a40079a046d0644bd0b1cd99f476bb9
25421 F20101211_AACBNG merrill_l_Page_39.QC.jpg
af0b7fdbb5b5b71faae43413c0d06c28
f3b20d47b6836fbc04ec15c805e060b1b5835582
84661 F20101211_AACBDL merrill_l_Page_15.jpg
13de4d6730e45f31d343141b9ba5da67
0e1539c22e6507d1cc52a59fc442a87e93ac1851
40746 F20101211_AACBIJ merrill_l_Page_04.jpg
f6f978121df68e88872e430fdb1c92f2
9addb0f4b1a850be34a0278ca139f1f358c2e0a7
2970 F20101211_AACBNH merrill_l_Page_53thm.jpg
60a9365c7008385c6cf442619cdf4a41
27eb1ede916318f1bf4ea555f71fef30ecc57218
F20101211_AACBDM merrill_l_Page_03.tif
0032c4a8caa84b4fceebcbe320b97472
038e46e342a1d61b779d5d03a4738be00158a5b0
49770 F20101211_AACBIK merrill_l_Page_05.jpg
a9a121dd9a0554c2afa44b8b25ccd2f1
3cf15d2dccf2514521dc3ba9f173842f21343156
27862 F20101211_AACBNI merrill_l_Page_12.QC.jpg
eda5a0aea67ede2c2a6bfd7f7a4e5173
770595234e017fcf0bf002dde71d24cd067103c9
68020 F20101211_AACBDN merrill_l_Page_65.pro
cf977fb39304be70ce51071b05c51ef8
2f8d6f01d7e13fbbb04da365280bd3dba473d3e8
51899 F20101211_AACBIL merrill_l_Page_06.jpg
57bf6181e70dc72a23da1a72a4d04fae
6ee68fdfcd735881ce06cb68530680abea691452
7534 F20101211_AACBNJ merrill_l_Page_55thm.jpg
4a6d1c0712fea670f054fa396a764e1b
5dae7c1654b6c4090a477e482550f472bf332133
F20101211_AACBDO merrill_l_Page_09.tif
7c25fce40440c9062fa3dfb98e9e4acc
cd4e4669a3e6df7ea28b069acb099f68efc33962
31798 F20101211_AACBIM merrill_l_Page_09.jpg
b9b572b826e450d6adf7132fd9891694
210e2952e443a04a5639605d49dda39d0190dd5a
25221 F20101211_AACBNK merrill_l_Page_24.QC.jpg
8af98d814b5e4ed8cf26647d72cba2b9
c8fc1500be346d3e4fe0b60c343ee91d23bf6516
24285 F20101211_AACBDP merrill_l_Page_01.jpg
b4301f3d56153b207d8c08c8c6b6b7fa
439eff22fb7c987efd63666436e13fd4976de282
85571 F20101211_AACBIN merrill_l_Page_18.jpg
23a96e07a4d46198b9be2b665d04044a
1f5e5e00d9b6729d83179fab03108f734486e166
7386 F20101211_AACBNL merrill_l_Page_13thm.jpg
79d307b371eb24a3b7fd597356b14ca8
22cb0e938df96a85011debb2c1dfcdf895d5dd25
86620 F20101211_AACBDQ merrill_l_Page_17.jpg
ad34319bb59668da43875b449107e86b
821bfd848d9d06e6153ac4d1fc1342a70078b2c3
28198 F20101211_AACBNM merrill_l_Page_47.QC.jpg
f7444213544b890bfa5340ee88b6c119
9215538731b712a8459fb7b7984544cceae8da5e
99487 F20101211_AACBDR merrill_l_Page_21.jp2
8164fa6450cdfae01f7985af6e60db6c
9a74c968fe9830f1a3ab382915f0ab5eb8c496bd
75995 F20101211_AACBIO merrill_l_Page_19.jpg
368b645f06016397a8d417fac62915b5
19ce69790131154d1214ab92780abaabfe8ad0b0
7020 F20101211_AACBNN merrill_l_Page_39thm.jpg
574aecc695ae23fabc08e709f5379e66
ba8b62136bf2dcbde500683b779817534534d355
1051974 F20101211_AACBDS merrill_l_Page_26.jp2
117da3dc307f2091b68ca9f3586a5e16
a8d3d998f6c4326a7298d5dd01da5c1c2bf43284
40797 F20101211_AACBIP merrill_l_Page_20.jpg
178898bcaa70175503a37e1a6e22e000
ea557bb3485d45d6003cad4f635fd2c2b2e9a531
3765 F20101211_AACBNO merrill_l_Page_63thm.jpg
fef6b00e1045984b49f9a7fac681ae1f
d4bbd7dfa4c1d322d3a2031a29db23541bd44a1b
1051916 F20101211_AACBDT merrill_l_Page_30.jp2
479f5217aa268cea711ca481dd0d9d38
34993cb0a090b7660c122c8206853ee215332c99
80432 F20101211_AACBIQ merrill_l_Page_26.jpg
56ee56f51346f58d9df1c1caf330b426
364f1c4dbeab2f76ed6eddb646b4e725622ce5e8
7590 F20101211_AACBNP merrill_l_Page_49.QC.jpg
1fc9aaf6a8f805eb4570e1232ca9e219
078653d6a17bec29c74c4c9bc1cd806bf7de777a
F20101211_AACBDU merrill_l_Page_17.tif
996021557bff49c371f80f9ad8157cae
a2038ef6b1128ffa04688d601319a02ef77c5f23
81190 F20101211_AACBIR merrill_l_Page_27.jpg
60c013fc3e897844887e7cc1d5a47176
732dd0eb76356eecca50ee9eaa1665e4878571cf
2385 F20101211_AACBNQ merrill_l_Page_49thm.jpg
d89af6f9775a44d741a54e96a9e711a9
5fa55dedd3c33361d58745a63378193a7139a020
2208 F20101211_AACBDV merrill_l_Page_56.txt
e4483d21310fda28a419fc3c5408bbab
45b1735278808e966b9fb3171138f92866637f44
80084 F20101211_AACBIS merrill_l_Page_30.jpg
9e142f0e06d0cb573d8d5b0552dd82bc
388fd6122323b30a12348e0adc3b4372923aaf61
2519 F20101211_AACBNR merrill_l_Page_33thm.jpg
554245f5679603786df08e50124edd1e
855bf0317a0e4f7b6ba9081726fc1f427c3e0aa5
3101 F20101211_AACBDW merrill_l_Page_09thm.jpg
59b5acb92bc0c34cfb2a0dd93337df2a
7c5c811c401e95af01f1b997ae358cd8e8514d5d
41803 F20101211_AACBIT merrill_l_Page_36.jpg
852a990d9e39de7976c9188a2a034a59
4e91d72c1ae2ea543cb10952af50a56f5b62f022
7592 F20101211_AACBNS merrill_l_Page_33.QC.jpg
4990fea365dc98d5d2e36045579f2736
3dc74fb3eafe585f6db3d00df99c1be42eb9e972
24826 F20101211_AACBDX merrill_l_Page_37.QC.jpg
daf5d7206946e1003ca0746c5c87190d
4e3455b123a8e05e5d002f157c2d4ca3e13bd902
75877 F20101211_AACBIU merrill_l_Page_38.jpg
4a4d389ddbb08079f2e76235e19581c7
c81379f7ebf3dacf7c2f43fd3191ff2b416e6297
1051984 F20101211_AACBDY merrill_l_Page_44.jp2
b2424455daf7ace160946e5288b5077b
c9513da0f0abd216281d6277c00eb85e060ed71e
78311 F20101211_AACBIV merrill_l_Page_39.jpg
11637f92562236026283d7ccbb982430
f015b44fc1872e6ad215d23831269a23d9adfd36
6867 F20101211_AACBBA merrill_l_Page_19thm.jpg
22369714a5e43f145cf39cb680c72d77
1bdd853b503e8fb3856d150fe7ee923900da312e
6932 F20101211_AACBNT merrill_l_Page_37thm.jpg
fe6d6416925c7147dc3c3f8f38d4056c
ed007152ffee9fec2e6011c74cfbb539579ff8f9
51873 F20101211_AACBDZ merrill_l_Page_23.pro
081519741622f208d242d6b0bc8cf5cb
2ff7dfdc46fecbc8b15a4dfca2673f471a3aff6b
79444 F20101211_AACBIW merrill_l_Page_43.jpg
137cbe245f7e2175856702785ba52fd7
5b9a389701273b4ce77994f4ac6d8a0a143cc171
23101 F20101211_AACAZA merrill_l_Page_49.jpg
2013d0ce896a3043f0d80efebd702152
733ca014d6dc632f664a12543a57891466d4981d
F20101211_AACBBB merrill_l_Page_23.tif
9cf8dcf322a343913df38ec5a4f5be49
2c00b3693e7a495289fe3aa8350864428be0c690
26634 F20101211_AACBNU merrill_l_Page_46.QC.jpg
b62b3f0127ddf4f355b6f7f86a30daae
7ff94a55ebc634a7ab351dd0777ae6551faf497c
87266 F20101211_AACBIX merrill_l_Page_45.jpg
37cd85d09896f1240b61b499a049623a
16ca207502e1afc1c6ebe76b94987b6a0231e465
55943 F20101211_AACAZB merrill_l_Page_57.pro
155cc20893f342d7d298ee34d12d158f
367149b3f07d380f8602ac60839f471415e17f2c
F20101211_AACBBC merrill_l_Page_62.jp2
fbf8d5e05047bb0965df2c1497d86251
f5d95aefd649257659599a28de0f4b99cd1dc699
7518 F20101211_AACBNV merrill_l_Page_50.QC.jpg
c4f1980ade74a4dec6443a516d2038f2
8ea71a43222ee6c821c78d49f919ed1b7f7824ab
F20101211_AACBGA merrill_l_Page_47.tif
6fccc5826e90f4da5950d3f4207ee816
9284d0267bf53b1b09ad99271c917d1c27ec69d0
84302 F20101211_AACBIY merrill_l_Page_46.jpg
ffacaea23acfc25db2d65e1afa830b59
d61445dc69dec0855fd8cda22013c04a4c13dbf5
F20101211_AACAZC merrill_l_Page_25.tif
abb46706fdaea62551b54e20c484cde7
293c701e20e87e3c6d61eef7dcffdabff9022922
51839 F20101211_AACBBD merrill_l_Page_44.pro
d8077ff129c6fb12fd052111b84dd685
445455d8f4aeb13efcc3ce4cb235b5d599f570b1
2309 F20101211_AACBNW merrill_l_Page_66thm.jpg
3c756c07499dd3c2ea18def70e04f098
f2ccad3d149446ad0b7c7866418878521f9435c2
14427 F20101211_AACBGB merrill_l_Page_52.QC.jpg
4b06a979b46babc83f604bba201cfe40
498b72efd87eaf4c79ef09985a3f4cb9bc707454
87863 F20101211_AACBIZ merrill_l_Page_47.jpg
287d9a3f5c12d6eab2c1652282623bdf
c1849f19424bbd536b72b96a8ffa754cbd95cff0
4177 F20101211_AACAZD merrill_l_Page_05thm.jpg
0b63e291ecb2d61db3b00cb6f2d83fef
25a226bb0b279b98ecd71d7b0f14887325dc408f
29361 F20101211_AACBBE merrill_l_Page_65.QC.jpg
f97719c0ce2bf4de77a74ca2c9bbb087
06ba834e29eb3f2124834c5d5fa7888440c8e922
11704 F20101211_AACBNX merrill_l_Page_36.QC.jpg
ccefb854aa3cc66c171014ea049fb340
fd3ba68ac393a54e82bb9130dd95042f750ff275
746481 F20101211_AACBGC merrill_l_Page_61.jp2
9de0898de577e353625f6f5223bb01f5
7a3313de08f41c45130563fce6f59ed5dfd222fc
F20101211_AACAZE merrill_l_Page_54.tif
efd83e4934cf1031ee85bef02912679d
ce3dac5b01de172cfb70ce1cbfc2db7e432f0c8d
1051972 F20101211_AACBBF merrill_l_Page_40.jp2
989f4ded7c4ce8f60a234c89202d1f5d
654ab6e79ce039d1d654d723a11cc9dfe2cc9959
7498 F20101211_AACBNY merrill_l_Page_14thm.jpg
235ce945d7d787c95c61cc9b9d626822
b275e8c12edc15875a18d295076d5df0a7ed6d7b
27288 F20101211_AACBGD merrill_l_Page_17.QC.jpg
8aa1389a3f3b5aa16217f03ef14db213
abd5cff4c3d30b094d42af19b1ece9006ec7c036
517 F20101211_AACAZF merrill_l_Page_32.txt
561af513419b42f9c9b083cba8cb2b39
43051a17a51a2461e5ababe7c6c69df8a587aeb9
2403 F20101211_AACBBG merrill_l_Page_50thm.jpg
a0dc0c6d141fc9046a311a15fcedef8c
2546f833f7df626db0da56214722a9ac2f00b5b1
45551 F20101211_AACBLA merrill_l_Page_08.pro
7483c3ce3212d28cc79bde100bcedc0f
cb70571ea30103f4f4cc611f72f23d26da0e6383
101370 F20101211_AACBNZ UFE0021558_00001.xml
aec446fb530ab72d4332fc37f259a116
3d5021e51222f45f323d3c06a738793c6be741d0
312488 F20101211_AACBGE merrill_l_Page_35.jp2
8770359bea18cf722ed65baa0093a8f8
9617b1aff0b2e9ce2315d520502ba1b471a04e40
7124 F20101211_AACAZG merrill_l_Page_16thm.jpg
f0e10dde90a15ff5d336697ab92f9745
72b41a9d484fb64edb58817c5767a58f4c9d58b3
16389 F20101211_AACBLB merrill_l_Page_09.pro
5df352cd9db69c8b55e78c7423c31a46
bd318b8fa5acf9d4237eec0169aee229e5f27af1
2049 F20101211_AACBGF merrill_l_Page_30.txt
a82428f5178bd6ee65c8e87b8f8ef2b0
41dfdbed670a26f0c38485bec6d3f1c9aa12b199
F20101211_AACAZH merrill_l_Page_42.txt
c3af6abb573d4411fb969d9a39485062
e6f267aad709369d347a7e3883d8af56b880c328
F20101211_AACBBH merrill_l_Page_15.jp2
b7d4b41deeb0e3978e9e23a9ed31a092
3fd82dc818fcb2d26ce6782a07ede01feeaf655c
51410 F20101211_AACBLC merrill_l_Page_10.pro
f332d99692e96c033f4cfc23a43a9e24
4077457869b39938d06c6e51b3b5117029a4b0ae
251657 F20101211_AACBGG merrill_l.pdf
3e0a31ae0d81d1988c4dea3d61c853ae
57d7d05ca7cc7ad75b3577d0f1522af4c5dac06f
7451 F20101211_AACAZI merrill_l_Page_54thm.jpg
cb41f74cc11733392e8cba7e8b3d137c
a270ae7e254c319c0422d1784a9df44d71f6508b
49190 F20101211_AACBBI merrill_l_Page_52.jpg
655fd44988d62a8f6bf7774a22ac06b2
670374ee459e0362f515cc3effe65435404be427
55341 F20101211_AACBLD merrill_l_Page_11.pro
72e9cf9e99eb063c5b9e66e12f84daf2
52039e1c92bf7049e4edd572935dfc436a0ccebd
1521 F20101211_AACBGH merrill_l_Page_60.txt
dcea9529fa1d691dca91c3fbe1243b97
5922835bb202cc7d9d8f01bb726fc03891e09eb1
28369 F20101211_AACAZJ merrill_l_Page_55.QC.jpg
10d78adc09c1a6c15526d985b69d7c4d
4338ea9ae20da6f2dd01bfde36720bb0f2b39cd0
6914 F20101211_AACBBJ merrill_l_Page_24thm.jpg
614dd4e27b5795eea79dc2b37e283bfc
b97d444e820ecd08293b94db427a9aaa42d754a9
55337 F20101211_AACBLE merrill_l_Page_13.pro
d8f8e170fe001540a2b993df7ebd77cf
9a08ad4038567603872a26d2f7f7452ecb24c1ca
F20101211_AACBGI merrill_l_Page_66.tif
57dd5010ec97d792e429d87e81baa8f1
43c38f30a5b1b37caa964cbd014479f4fb0e3fdf
48313 F20101211_AACAZK merrill_l_Page_39.pro
da026d7d8f95705e5026cbf7785a602e
2933f018064db3693f103df2b92676b0be87a39a
2262 F20101211_AACBBK merrill_l_Page_12.txt
520b044af0d75853166898c586d3a036
d54469639e57271e12ca6213242348cb4403ec8e
52980 F20101211_AACBLF merrill_l_Page_15.pro
00c4d1b0ead76f88de60d010170a8dfc
25f2c21796fe3dc68de77d0fc65ffbb4b360eebe
2122 F20101211_AACBGJ merrill_l_Page_15.txt
1482cde24d7bf66621bb4edf516280e0
2e57143050d8688ad9458693042f701e6aaf5e72
77720 F20101211_AACAZL merrill_l_Page_37.jpg
d1c64146d6665b3ffdaad5df16fc51cf
b3550b0495ea08d7160ee9be9ba0c1ebaa6f73ff
F20101211_AACBBL merrill_l_Page_05.tif
7f5ef3fd6da23546cb784224709770cc
548858653212c54ce50f9d3293948744d3c0a5dd
53969 F20101211_AACBLG merrill_l_Page_16.pro
8c4f836da97d99c72896a578607f20ee
e3fd92181c4ace927558533b6dc6dd7ab2b7355a
50821 F20101211_AACBGK merrill_l_Page_30.pro
b678e7d44ee20247dd3c40906f49094f
42673b325e15093e8d085422a71653f0fc2c21d6
1051938 F20101211_AACAZM merrill_l_Page_24.jp2
24b95a55c622f2bded79f769c6dfd43d
b9b30e996ef57c39efaf1156aeac93ba50575320
23721 F20101211_AACBBM merrill_l_Page_33.jpg
cc350a23b897823267a0c747ded79035
460b3aa2504b73e9c5447e35504c88076731b1d3
47750 F20101211_AACBLH merrill_l_Page_19.pro
92233a8d8029e55f3f796242298395c2
bd6faa9ef8e756a1cd216257f456fc32a5723a41
76 F20101211_AACBGL merrill_l_Page_03.txt
7fbc6c17c8fa1c40894e38591fb65ddc
cf5b398b56f7bfd0668e157d2b3cea463e434527
F20101211_AACAZN merrill_l_Page_38.tif
5446482cbce44190390e215495b7d457
252962cb5e2610f3aad2627854d6676b6fe1d7a8
7908 F20101211_AACBBN merrill_l_Page_01.pro
9b4e1b2a8a86a96f55cdcfe60a045d94
e52ea33daa8297553b80aa19f3a6126ad61cbb37
17827 F20101211_AACBLI merrill_l_Page_20.pro
b4c52f00c4aa620a9a1a90852deffb04
090ae2f753d42da2667b3590622200cd22149cad
21636 F20101211_AACAZO merrill_l_Page_04.pro
c801d5c566530f19a74e31e549534bb5
4a13a5098ffa1e5faf27eb5db6ab749c3a8ef148
87975 F20101211_AACBBO merrill_l_Page_14.jpg
f9290aafa2092c1b6581fecd5e40a1bf
bedb858e83db6531957d2f9a9d143506755a1e51
44818 F20101211_AACBLJ merrill_l_Page_21.pro
177e7bbb2d18a0ab3a163e74d22a2abb
d4396fc5e1ce889de2a9eb2b9c554875ef392f3c
7367 F20101211_AACBGM merrill_l_Page_17thm.jpg
26fa9e2899c487d48909436d1efa8c2b
495236ed9fc856f813cddb347501f7a54cc50242
24478 F20101211_AACAZP merrill_l_Page_29.QC.jpg
4bc85d20b34b5aac9a61ca6eb469c4f5
f5bc94026ffa48d59bb45ab920ca96914ab403d9
1336 F20101211_AACBBP merrill_l_Page_02thm.jpg
95aafec7e401e07f1c9d0d01107ff2a8
3216af2b8b9c1723fc0528e6540e420c176c0741
51233 F20101211_AACBLK merrill_l_Page_25.pro
5606145f0d7f3763d825be5ec13b36c8
f93b7df95a818cbffd1c28bacc98728e18feb72f
7139 F20101211_AACBGN merrill_l_Page_64thm.jpg
9ca63bfe28e98cd8653ec608ac194abc
7568e4ab0f2fa7306a6f65bc9f33b92d4b257b29
27490 F20101211_AACAZQ merrill_l_Page_16.QC.jpg
8041c9658171b4453ac45b9fa8be663e
ec99822f94ca4457a1634ca7951720b61c2fba44
F20101211_AACBBQ merrill_l_Page_30.tif
3520bf50e4c079387d92dfe678de2c75
9a20349c18203b2d043a2b839df82b6863796c58
48224 F20101211_AACBLL merrill_l_Page_29.pro
bc6ddf2289f3d99b94d67560c2ffb712
36c68e59a36dd8e852b07f7fb7da1fd211c99069
F20101211_AACBGO merrill_l_Page_50.tif
128a25e4f7f196cc8537be4643719d03
bc0f3084b9e029426da1dc15d0be5f86ff7fcf0c
104635 F20101211_AACAZR merrill_l_Page_65.jpg
0996439d8a88975033544e0e3ae88061
e3461671abb88caac8bf50d25978303e14b28382
52967 F20101211_AACBBR merrill_l_Page_42.pro
accaf53ef06373cbf4c96085870876ad
0b3432da37f0a14263b603a111c4dd34e7ca1725
12969 F20101211_AACBLM merrill_l_Page_32.pro
8505a802d313049ae26bcea9df34ede3
fd8b496bcadf2b5addd1ca81aa98705f2a6696de
56365 F20101211_AACBGP merrill_l_Page_56.pro
21d8e1c43fa1dd54a6958c7c66e758eb
35bf679933aa6a69c6eff5efdf0bbe1a1502928e
2039 F20101211_AACAZS merrill_l_Page_37.txt
a327fa6f78359f51b501d9a360f830bf
4979341dee6ca40727ae013ec4bc78e894d6dc48
F20101211_AACBBS merrill_l_Page_46.tif
688d9525ef35a33d17373aae3eadb7fb
c268e240ffbd80be9e330eddae2c43bff66f0e1b
48264 F20101211_AACBLN merrill_l_Page_37.pro
0cc73033525755d3dd8f1b0a278172de
9864a789be48e0c76a57aae19ad9f300ba8fc636
1906 F20101211_AACBGQ merrill_l_Page_39.txt
9b69fe935e1ad0ba937625588e305dbb
8a292a5a344ad34caa0538d73590e22076b724f9
8640 F20101211_AACAZT merrill_l_Page_49.pro
7eb3724bad5570fdde159da62335f2bf
fcf09ea01d44161582fe4675aee9f5ebd4545cc6
1051908 F20101211_AACBBT merrill_l_Page_25.jp2
caaddec594ee9f3d85b853a6a5615d23
baa590e456730569170de3ba41ab754b3328ba18
52797 F20101211_AACBLO merrill_l_Page_40.pro
2c4a779cbc3ecd77a282960e61bdd6bb
15a3f50f283402bc50b5f31e9dc0dd6e73ad0fe7
F20101211_AACBGR merrill_l_Page_56thm.jpg
b83752004c9721a80606fa580f41b384
d951a418bdc60740c877c236aab8c214729a6a09
520 F20101211_AACAZU merrill_l_Page_51.txt
349dbd6113d55fcd00a85b73c94007c1
d418e32361510800354b30f8b304493d37688033
4784 F20101211_AACBBU merrill_l_Page_48thm.jpg
e3401af555306ab7fb287d2a4aac0881
5158164f88104c31fcb2a6cf1b0a8968d46bf973
49197 F20101211_AACBLP merrill_l_Page_43.pro
bc223011139de9ea3e2e5e1866794016
436bfbee13d46585663b7aa05c6f7906e3992b10
52712 F20101211_AACBGS merrill_l_Page_59.pro
72ccf26016454d078ed6f1af36558970
6b5b7d713d7f35c3db7072e28753937e3f6ebfaf
1051941 F20101211_AACAZV merrill_l_Page_47.jp2
f7b49fa602f4845c924b7f3569bd095b
88071826983ec3a72d6004046fd35d57a25d2f68
F20101211_AACBBV merrill_l_Page_32.tif
e1d001195395b5aa9955ad5123eaff1c
4e78ef0656443ce58a19532bcdd0660f43a036b2
54332 F20101211_AACBLQ merrill_l_Page_45.pro
4e589bb70bf7233d2a6e6c2a6bac2128
03b04ec1b0ffbbfc006f7e9c092a32b22e05eadc
77172 F20101211_AACBGT merrill_l_Page_29.jpg
49377b0597181b79a742165f6ce471a2
96e53fe6bbd7bc89e1dde8477eeacb82418951a3
3566 F20101211_AACAZW merrill_l_Page_36thm.jpg
338f873a3a854e289ada660e87057ded
67fb1b0465e15dc1bed93567b655e1ee325f6ea8
7286 F20101211_AACBBW merrill_l_Page_44thm.jpg
3e43d24e6dd0ee14906b5e7065de099f
7655806cc010858013dd5f6ad131c0820c210a1c
31741 F20101211_AACBGU merrill_l_Page_06.pro
0bb894217945fa32ebdef7d53d73e462
412548b0cb0a61f6ca4fac7af1a3252465e7787f
27665 F20101211_AACAZX merrill_l_Page_54.QC.jpg
575774d9a18f704f89573595751fe4f4
3be0e44a7963c84998894abdd4f9f9aab820764f
25944 F20101211_AACBBX merrill_l_Page_10.QC.jpg
517cb6c3a2ee00ee4b296f83394ef990
82f19ad6a0057fdf7c8afe125d74dab9c5e0c7bd
53015 F20101211_AACBLR merrill_l_Page_46.pro
436f8cf3b3bad2b74391bb9c9592612e
ff8e78d4609272ffc2a623b78fb75bb027be4fd0
83968 F20101211_AACBGV merrill_l_Page_10.jpg
066d7fe7c928c712dadf7f3a13e9f1b5
d973d70dea820bdaa3f5c7950b890473168f877f
27131 F20101211_AACAZY merrill_l_Page_22.QC.jpg
b1f2bdc05ca2e83d66555d545acb94c1
b895c8037dc79d9e519048613f7facdc799b264a
7293 F20101211_AACBBY merrill_l_Page_15thm.jpg
39387523e053df769a18b111efcca428
c62302cd686e3d2331b44e3997ae1e1b048170a5
8628 F20101211_AACBLS merrill_l_Page_50.pro
c5a16b2f148bf4ff8c2f8fbef91105a2
4098983d51d6d64e47d5261464badbbb8691b059
F20101211_AACBGW merrill_l_Page_55.jp2
02e1cbf68b8c4e99aa9ee263a825f5f3
ed3e37c1af1f112c375e9d0aeb0c747c6602bccd
1051957 F20101211_AACAZZ merrill_l_Page_42.jp2
d0cedb577a1dab28c44d8ba05880dd0a
ad6dca068a122f7e1c0de79f1b308f35e5ccc3fb
10278 F20101211_AACBBZ merrill_l_Page_09.QC.jpg
7278ff4893f248ae7cbac5d937c7ce40
27344a72ba264fbe44b997684b791bd337e5f750
21092 F20101211_AACBLT merrill_l_Page_52.pro
23c31d07edae39cdca9cdba94364aa7b
b1e2695e7fe19c8dfb0a0ff4d8de6689ab4efe2e
212932 F20101211_AACBGX merrill_l_Page_49.jp2
89bf2e17a3a4616c9c9e9fe70af0263b
8e55febb0a3d0dff1fdc099ac1d9c32387f6d1a2
9101 F20101211_AACBLU merrill_l_Page_53.pro
cce326292cda6c043306f42a0c723e68
1ab85c1ad545bfb8e8382f668df6e807d8b6d900
F20101211_AACBGY merrill_l_Page_60.tif
de2183016a3c44cd2a68a17a43d58e23
5d30e7a73dc7309afc625aed52c662f74c95f1b1
55191 F20101211_AACBLV merrill_l_Page_54.pro
1447311a14f52b7407615bd84839f48e
e913645992007719c7e4fe3d65eea0a4875e2e05
F20101211_AACBEA merrill_l_Page_23.jp2
e4d6ee3f0f5b69304199ba320279f01c
90028037f1f212ca4899db33ed3079758ce03c4e
86915 F20101211_AACBGZ merrill_l_Page_13.jpg
a464d8500e1d2df8ee467c56afd01258
a8a663f707bac4f7ab70dfee32f7ab536af01b88
37227 F20101211_AACBLW merrill_l_Page_60.pro
f3d1276c89562bea3d9bb64e4922c59d
01f9280448c22c8f8f5d6d2bb71657b7330aa93a
50694 F20101211_AACBEB merrill_l_Page_27.pro
4cd4da9e0ea21ca85a3ead620faa2087
deb24d3f1df1b2a19a1e7f8ac5ab3181ea4791d8
58738 F20101211_AACBLX merrill_l_Page_62.pro
de01e2d27dc91f370ff75539f5c30cd3
e046765e9510021be00bbdb2525730ad887f4ee0
F20101211_AACBEC merrill_l_Page_35.tif
acfbf37357ac1c7947a0e3ede2400d62
bdc3c9ed61ebba335ec12f1f7ad5447634ef6c25
53922 F20101211_AACBJA merrill_l_Page_48.jpg
dcf2989e2bbe625cb2ab365b598007c0
7eb2eecc1f8071dd8c1f03a5851aead9643d4b43
22022 F20101211_AACBLY merrill_l_Page_63.pro
fe3e9d715f091afc8470f8d0bf485920
b25d82f3adccac94180c268c0b1d444cea6802ed
27405 F20101211_AACBED merrill_l_Page_45.QC.jpg
45f2eefcd0265c5ce41d62b2286637c5
85e1afaf06638980f696b0df5a7a5ed53360b6e9
23048 F20101211_AACBJB merrill_l_Page_50.jpg
cbbbbb15be387c274a63f913bb2bc5f3
73a2c6e0d6565e255127a777dd9f0d8971ecff31
57117 F20101211_AACBLZ merrill_l_Page_64.pro
26ad7e741c76f572c8d2f748eafacafc
35bf48cdb69d2ea5cb901f0e15db73298345bc8c
2001 F20101211_AACBEE merrill_l_Page_24.txt
b184f353627543c3c03cd19555d45616
f54f94006b25328605c3a786c392b9d17d031750
29505 F20101211_AACBJC merrill_l_Page_51.jpg
9f008473887215da9a64bd59c1bae47e
398894d08fe85b7dfadbcee3a532995fc308571b
90309 F20101211_AACBEF merrill_l_Page_62.jpg
455d66dab6bbc91ac1694cf48fb82cab
c9833cc6fe195fd9902be6d53fa926ef5de758df
25041 F20101211_AACAXH merrill_l_Page_30.QC.jpg
10002aa9579a6d5cb092887aa459fcf3
37bddbdfda803d583ee27651192cc35b94653d75
7174 F20101211_AACBOA merrill_l_Page_01.QC.jpg
810e67482eb4e29e15a1832bcad57821
65d4bca14d44edf31788ddb7d22e1021fd175eb0
89805 F20101211_AACBJD merrill_l_Page_55.jpg
6fec5a63981e1dfbb169903f313054ab
13db76501ccaca4a4d92e92c16670c6ac4ddcafc
F20101211_AACBEG merrill_l_Page_48.tif
081fb5eadfff105199f880f01864bc56
3dbc1fef2501ece7ec7146d3d24e84dbb33ad6c3
23864 F20101211_AACAXI merrill_l_Page_38.QC.jpg
3c1e24791de2515441fa53706e1c90b6
a6d920f71eb91e96ea60a0d2599bc9611ddf0264
3046 F20101211_AACBOB merrill_l_Page_02.QC.jpg
dc82c8a434b9ff2d6af4ecadd6643c0b
3b2b29e9394f170feaa00f12fe071f7b41915e11
64219 F20101211_AACBJE merrill_l_Page_60.jpg
756deb07a1200ed24df490a86250fa49
41ae8c60ee6db8dfc0628c6f213521dd3134199a
54659 F20101211_AACBEH merrill_l_Page_22.pro
bf8f4ce4de6adfc877081c2f3106a470
867a45ee6b7b66edce0a5f76c16d37056c1c0f03
27221 F20101211_AACAXJ merrill_l_Page_11.QC.jpg
4590ce10fc5529e6d73a47e23eaee9d1
d2b94a3a16ce7a0359ba3c3fa6cd9775e3565a77
3929 F20101211_AACBOC merrill_l_Page_06thm.jpg
dd470bb439339e9d4a65198a91d282a3
849e10d1a7c4f8dcde70e013029f3879f70d961f
38806 F20101211_AACBJF merrill_l_Page_63.jpg
d04cf24aaa55aa6c59009e2a43211bd9
727938d2e50ded256fda8579ba1dbe5741a8837b
3933 F20101211_AACBEI merrill_l_Page_20thm.jpg
64e40a4e2fc38c20a0abcf5d027464e0
7afae2c0862dd65c6aa1d02d561203d7adeeb60a
F20101211_AACAXK merrill_l_Page_39.tif
aef2f3fd6e0cd6bb0e49efaf3c0a9ec5
9e982f5c494bfd4dcb30d3b1a60f8820a8d56f5e
2413 F20101211_AACBOD merrill_l_Page_07thm.jpg
c3ad30f5439384fa51e25f09ef35b9da
5f6517d14f92a90ffb8c4ea31b7600161e2cc291
91271 F20101211_AACBJG merrill_l_Page_64.jpg
c70529e346148786cbf4da77efe5799c
969a5a38b2230cf3cb3609758f8daf690804f9fd
1051912 F20101211_AACBEJ merrill_l_Page_16.jp2
5626881464c6dd3d2d589309d0fd663a
be24641a3cb6b98df24bd6fbf5edc465b3bdb16b
2192 F20101211_AACAXL merrill_l_Page_01thm.jpg
d3963190d5dc372a45f93638709ebda4
42d38c31e2ecfa7134c031b757b64bccf530dff0
23580 F20101211_AACBOE merrill_l_Page_08.QC.jpg
a095f4551faebe3e65b19816798a0c24
cfb41b1562cdc46e4ce962536c26e8e5a4b92cc1
23451 F20101211_AACBJH merrill_l_Page_66.jpg
3be6554677b984bcf9efb88a54f748fe
761ad2c1b1052577a8341c3542954545fb11fd26
1946 F20101211_AACAXM merrill_l_Page_43.txt
64032ad2330d3799afff87b43771c588
b64e53fa9c59b57045a1a68e27cc5fdb1759ef86
6312 F20101211_AACBOF merrill_l_Page_08thm.jpg
8fc502098f1cd524b20919d8fc3e40de
845e0fb46ad82ab8581d7ce1f6fa071ea8b69475
229628 F20101211_AACBJI merrill_l_Page_01.jp2
615336bfa483f4f5f9805e6f10e4fc94
97e25e0e10cd760a041b4379a548eab6c5cfaceb
46433 F20101211_AACBEK merrill_l_Page_38.pro
44aecbf2f96c4c97c6fae130285fd667
13f6812019893b4b0aa606fa200be9d9feedd9c1
F20101211_AACAXN merrill_l_Page_14.tif
c8ce1ce8ec82d72911a62a272a323d1b
16c856754a83e8148945a25643ffe527ef6fd352
7416 F20101211_AACBOG merrill_l_Page_12thm.jpg
e299d06cd0080ce1ea53293af5c7ae72
b02a75660e083a391b84fbb2ae787959698a24ed
23104 F20101211_AACBJJ merrill_l_Page_02.jp2
ee2e27a00729ea2c154f0eab86b43976
b107c4a87fc03887baaa0c1eee05d7cb46dde1ee
3063 F20101211_AACBEL merrill_l_Page_03.QC.jpg
a5851d0969203a3dd716968232407e4c
4a69625341d52831d3f6ea2751ffedb4446b0606
54683 F20101211_AACAXO merrill_l_Page_14.pro
5c5bece53bba6f1cd23b6198f9ff3b1d
b73758b53e3f6995fd16b0045b4d942830a1ed9b
28002 F20101211_AACBOH merrill_l_Page_13.QC.jpg
80e15908168e015802590277bc2f001d
11c0b4cdd2c0d6f759b670f62de381e22665bce6
F20101211_AACBJK merrill_l_Page_05.jp2
8c3461fca6648c065a788a5a283560e3
739bbc361020596a4b6e0dc83cad2fc2661d9088
28220 F20101211_AACBEM merrill_l_Page_57.QC.jpg
ec6b77cc2ea1f0a9052c20a4e1b45b6b
1ac9b61758b1ee59390f39215c7acafc71d1c51c
4230 F20101211_AACAXP merrill_l_Page_52thm.jpg
be9594216d7e25795fd361f8f027bfb6
3825da2b5256e46099503c002253600a49ff94fd
27899 F20101211_AACBOI merrill_l_Page_14.QC.jpg
b2590058222f20605edb13eeb1729345
37f906c2a08b13b3e4d405770bccbebe9e6bdf0b
379162 F20101211_AACBJL merrill_l_Page_09.jp2
fc47909302d1e2bff5860199cbf41b9c
0b8de1062ebca76ff1edf527ca4bb001976281ed
88369 F20101211_AACBEN merrill_l_Page_56.jpg
134c096bffe7cb9afc4858e2fe61c945
fdedd09be3e0f094cfeab2f6e9e716243e03337e
1051982 F20101211_AACAXQ merrill_l_Page_28.jp2
3a3e267cdd98e6cb8a40157d52a0f4ed
81dd20ac6fb3cc192ea0b3229855a76bb21dcbe0
26852 F20101211_AACBOJ merrill_l_Page_15.QC.jpg
ecd2c7b81b3cf76bb746caab8a54a46d
85a9fe3790108e55ebe0554c55f04ed7588fdd04
1051965 F20101211_AACBJM merrill_l_Page_13.jp2
fa057774b919187a7de2983647dcdd30
7f655fac5224ae74351753cd12f814fbe6f0c377
6308 F20101211_AACBEO merrill_l_Page_21thm.jpg
cd48c5da39fafe2e1fb8e38e21b64c6e
824b228071eb70e4a29b6455949b4567adac930d
1034 F20101211_AACAXR merrill_l_Page_52.txt
7ded3c82a5f150d731df23a09da1461c
3c5aa8f83f3b91e75ac3ae52f6750af10cdfe2d2
27138 F20101211_AACBOK merrill_l_Page_18.QC.jpg
a2ec6f383fd9b9245e9a02088b1ecefe
4c04e376c14424e8bd1426e1199b298984288d76
1051931 F20101211_AACBJN merrill_l_Page_17.jp2
4362e1667e1f0016edabbdbbd068f335
7dd6f9510cad964bb108901eade239b409d2b8cd
2044 F20101211_AACBEP merrill_l_Page_23.txt
5f6c1b024248672e03f01c0d6248947c
644ef20a72c79cb3f65ef33363febd50464da336
7204 F20101211_AACAXS merrill_l_Page_18thm.jpg
359118049d80253394b414a9e84e3080
d147a937cfed6156406526155bcc518fdaae09a5
24590 F20101211_AACBOL merrill_l_Page_19.QC.jpg
594989df2d6ef56c85a19926f9c50886
1b7089a7cea27485c2338a86ea83123f1ea4da6f
F20101211_AACBJO merrill_l_Page_19.jp2
98ce017154db5ca765c483bf3796592f
c1fdb6a1e6e673cc4cb2255d71f83051580c9cce
7120 F20101211_AACBEQ merrill_l_Page_62thm.jpg
e7240bc89487692bbafe3b135b98e45e
94a0f8b9a14c3e9f5b864e6f81f82b76f1385d74
26937 F20101211_AACAXT merrill_l_Page_32.jpg
79dd7998e8c0f4ab5776167e9c407e91
8e98b8b7f73df851fc731ac42ef305b73c4c1204
7251 F20101211_AACBOM merrill_l_Page_22thm.jpg
341e80b30bdfaf202a22d2c9ac450bb1
450a92186cb14f253be7cb99e622a894a8313ddc
9072 F20101211_AACBER merrill_l_Page_34.pro
e6a383d2889acbc3c876cf3a98136d31
d93cd3c79600a8c4c1e4e6731728ec55fbc1eaad
19042 F20101211_AACAXU merrill_l_Page_36.pro
a2de8c3e3196d4e1d3f07d1327165d59
741ba986bb80ff850b378137eedab02afab93741







INFLUENCES OF MEDIATED PRIMING ON
TIP-OF-THE-TONGUE INCIDENCE AND RESOLUTION


















By

LISA A. MERRILL


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2007








































O 2007 Lisa A. Merrill
































To my parents









ACKNOWLEDGE1VENTS

This proj ect would not have been possible without the support of many people. First, I

thank my advisor, Dr. Lise Abrams, for not only her continuous help in refining my ideas,

developing stimuli, and revising the paper but also her incessant encouragement and motivation

that kept me enthused and optimistic about my work. I thank my supervisory committee, Dr.

Wind Cowles and Dr. Ira Fischler, for their time and effort to help me successfully and

professionally complete this proj ect, and their thoughtful encouragement through their kind

words. I thank Kati Brown and Sara Fernandez for their generous help and humor during stimuli

development. Finally, I thank Will, Julie, David, Matt, and my parents for always providing me

with unconditional love and support in my decisions and pursuits.












TABLE OF CONTENTS


ACKNOWLED GEMENT S .............. ...........4.............. ....


LIST OF TABLES ............... ............6... ......... ....


LIST OF FIGURES ............... ...........7.............. ....


AB STRAC T ......_ _.............. ............8... ....


CHAPTER


1 INTRODUCTION ............... ...........10...................


2 EXPERIMENT 1............ ............21......


Method ............... ............21.. ...............
Re sults .............. ............26.. ...............
Discussion ............... ............30.. ...............


3 EXPERIMENT 2 ............... ............37.. .......... ....


Method ............... ............37.. ...............
Re sults .............. ............38.. ...............
Discussion ............... ............44.. ...............


4 GENERAL DISCUS SION ............... ...........54............. ....


APPENDIX


A Targets, Dominant Names of Prime Pictures, Secondary Names of Prime Pictures,
and Unrelated Control Pictures ............... ............60.. ..............


B Experiment Instructions ............... ...............62........... ....


REFERENCE LIST ............... ............64.. ...............


BIOGRAPHICAL SKETCH ............... ...........66...................










LIST OF TABLES


Table Page

2-1 Initial responses (in %) following production of the prime pictures'
non-secondary name and control pictures' name in Experiment 1 ..........._. .........33

2-2 Initial responses (in %) following production of the prime pictures' secondary name
and control pictures' name in Experiment 1 ................ ......... ............34

2-3 Initial responses (in %) following following prime pictures that shared the part of speech
(POS) with the target, prime pictures that differed in part of speech from the
target, and control pictures in Experiment 1 ................ ......... .............35

3-1 Target retrieval (in %) following production of the prime pictures'
non-secondary name and control pictures' name in Experiment 2. ................... ...49

3-2 Target retrieval (in %) following production of the prime pictures' secondary
name and control pictures' name in Experiment 2 ........._.._._ ............. .....50

3-3 Target retrieval (in %) following prime pictures that shared the part of speech
(POS) with the target, prime pictures that differed in part of speech from the
target, and control pictures in Experiment 2 ............... ............... ...._51










LIST OF FIGURES


Figure Page

1-1 Phonol ogi cally medi ated priming vi a spreading activati on ............... ...............20

2-1 Order of tasks for Experiment 1 ............... ............36.. .......... ..

3-1 Order of tasks for Experiment 2 ............... ............52.. .......... ..

3-2 Percent correct target retrieval following prime and control pictures for "TOT"
and "Unknown"/"N\one of the Above" responses in Experiment 2 ................... 53









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

INFLUENCES OF MEDIATED PRIMING ON
TIP-OF-THE-TONGUE INCIDENCE AND RESOLUTION

By

Lisa A. Merrill

December 2007

Chair: Lise Abrams
Major: Psychology

Previous research indicates that tip-of-the-tongue (TOT) states, characterized by the

temporary inability to retrieve a known word, can be reduced as well as resolved via

phonological priming, i.e., presentation of prime words phonologically related to the TOT target

word. The current study investigated the influences of mediated phonological priming, in which

the prime and target were indirectly related, on the incidence and resolution of TOT states.

Participants answered general knowledge questions designed to induce a TOT state and named

pictures in which one was either a prime or unrelated picture. The prime was a near-synonymous

picture that corresponded to a dominant (e.g., motorcycle) and secondary (e.g., bike) name,

where the dominant name referred to the name typically produced upon seeing the picture. The

secondary name was phonologically related to the TOT word by sharing its first syllable (e.g.,

biopsy). In Experiment 1, participants named a near-synonymous picture before answering a

general knowledge question, and the results showed that producing the dominant name of prime

pictures increased target retrieval and decreased TOT states for subsequently presented targets,

relative to naming unrelated pictures. In Experiment 2, participants named a near-synonymous

picture after answering the general knowledge questions, and for targets that were in TOT states,









naming the prime pictures' dominant name increased resolution of those TOT states compared to

unrelated pictures. Because production of the dominant name led to decreased TOT incidence

and increased TOT resolution than did an unrelated name, these results demonstrate that the

phonologically-related secondary name (without being produced) primed the target for retrieval.

In addition to providing another method of priming TOT states for future research, exposure to

the TOT word's first-syllable phonology via mediated priming may be a process involved in the

spontaneous resolution of TOT states in everyday life.









CHAPTER 1
INTTRODUCTION

Tip-of-the-tongue (TOT) states, characterized by the temporary inability to retrieve a

known word (e.g., Brown & McNeill, 1966), are common disruptions to fluent speech that are

not always immediately resolved (e.g., Burke, MacKay, Worthley, & Wade, 1991). The cause of

TOT states is thought to result from weakened phonological connections, which makes the TOT

word unable to receive sufficient activation for production (e.g., MacKay, 1987; MacKay &

Burke, 1990). Support for this view comes from previous research, in which presenting primes

that are phonologically related to the target word can facilitate TOT resolution (e.g., seeing

abdomen primes the TOT word abacus; Abrams & Rodriguez, 2005; Abrams, Trunk, & Merrill,

in press; Abrams, White, & Eitel, 2003; James & Burke, 2000; White & Abrams, 2002). This

method is thought to mimic the spontaneous resolution of TOTs in everyday life, as exposure

(i.e., via seeing or hearing) to a TOT word's phonological connections induces the word to be

more accessible but without a person's conscious knowledge (James & Burke, 2000). The current

experiments examined TOT incidence and resolution via phonologically-mediated priming,

which may be another means of spontaneous TOT resolution. That is, instead of direct exposure

to the first-syllable phonology of the TOT word, TOT incidence and resolution may also occur

because of mediated exposure to the first-syllable phonology of the TOT word, without actually

encountering that phonology directly.

Mediated Priming

Mediated priming involves primes and targets that are indirectly related by a connecting

link (e.g., pasture and milk are indirectly related by their connecting link cow) and is beneficial

for investigating factors of spreading activation. Spreading activation models contend that

activation of a lemma (i.e., the semantically and syntactically specified representation, which is










the word concept and corresponding properties) spreads to distant, indirectly-related and close,

directly-related associates, where relationship can be defined at multiple levels, such as meaning

(semantic) or sound (phonology; Balota & Lorch, 1986). For example, as seen in Figure 1-1,

because milk receives activation through pasture's directly-related semantic associate cow, milk

is an indirectly-related semantic associate of pa~sture. In addition to semantic associates,

mediated priming also occurs with phonological associates. For example, calculator is a

phonological, indirectly-related associate of pasture through the mediated prime cow. These

latter types of mediated primes were used in the present experiments to influence TOT incidence

and resolution.

Even though indirectly-related semantic and phonological associates are activated via

spreading activation, Dell and O'Seaghdha (1991, 1992) proposed that these lemmas receive

only a proportion of the original activation. In Figure 1, the connector lines between the lemmas

demonstrate this notion. For example, the fully activated prime pasture cannot send its full

amount of activation to its associate cow, as it sends some of its activation to other associates

(e.g., passage, grass, etc.). Of its received activation, the associate cow sends a share of its

activation to its associate milk. Even though indirectly-related associates receive only a

proportion of the original activation, they can nevertheless influence activity in the lexical

system (e.g., prime other words for activation). Thus, when considering mediated priming's role

in word retrieval, even though the prime is indirectly related to the target word, the target word

may still receive enough activation to decrease TOT incidence and increase TOT resolution.

Previous research has used mediated priming to investigate the processes underlying

word recognition, word retrieval, phonological processing, and lexical activation (e.g., Cutting &

Ferreira, 1999; Jeschniak & Schriefers, 1998; Peterson & Savoy, 1998; White & Abrams, 2004).









For example, Peterson and Savoy (1998) demonstrated the effects of mediated priming on word

recognition using near-synonyms, which are two or more words that correspond to the same

target. The most frequently used word is referred to as the dominant name, and the other, less

frequently used words) are the secondary names. For example, a long upholstered article of

furniture to seat more than one person can be appropriately referred to as either couch

(dominant) or sofa (secondary). Peterson and Savoy (1998) first determined the pictures'

dominant and secondary names by having participants name a series of pictures for which there

was more than one appropriate name. A picture was considered near-synonymous if participants

provided two basic-level names in at least 75% of the names given, and the frequency of the

names was distinctly disproportionate. The most frequently used name, averaged across all

responses, was the dominant name, which was produced a mean of 84% (range of 67% 98%),

and the less frequently used name was considered the secondary name.

To examine the effects of mediated priming on word recognition, participants viewed a

near-synonymous picture and an experimental word that was either phonologically related (e.g.,

soda) or unrelated (e.g., horse) to the secondary name (e.g., sofa), and they produced the

experimental word aloud. Participants named the phonologically-related word faster than the

phonologically-unrelated word, demonstrating that word recognition was facilitated via mediated

priming. Viewing a near-synonymous picture spread activation to both the corresponding

dominant and secondary names, which similarly activated their semantically- and

phonologically-related lemmas. Being phonologically related to the secondary name, the


1Peterson and Savoy (1998) described their task as representing both word recognition
and production. However, participants read aloud words presented on the screen; they did not
retrieve the word from memory, as the word was provided for them. Therefore, I refer to the task
in terms of word recognition only to differentiate it from tasks that require self-initiated
production.










experimental word received enough priming to be activated and produced more quickly, relative

to the phonologi cally-unrelated word. Another alternative explanation may be that instead of the

dominant name spreading activation to the secondary name, priming may have occurred only for

participants who silently named the picture with the secondary name, thereby directly activating

the phonology of the experimental word. However, these results of mediated priming have been

replicated in similar word recognition studies using Dutch words (Jeschniak & Schriefers, 1998)

and homonyms (Cutting & Ferreira, 1999), suggesting that mediated priming did occur in this

study .

Furthermore, White and Abrams (2004) demonstrated mediated priming in word retrieval

by using an associate word-stem completion task. Participants completed word-stems in one of

three conditions: 1) a semantic condition, in which a word was paired with the first letter of a

specific semantic associate (e.g., beach-s with sand as the target); 2) a phonological

condition, in which a word was paired with the first letter of its homophone' s semantic associate

(e.g., beech-s ), and 3) an unrelated condition, in which an orthographically similar word to

the homophones was paired with the first letter used in the other conditions (e.g., batch-s ).

White and Abrams (2004) found that relative to the unrelated condition, participants produced

the semantic associate more often in the phonological condition (i.e., sand was produced more

often following beech than batch). In this example, beech activates its phonological associate,

beach, which in turn activates its own semantic associates, such as sand. Thus, beech and sand

have a preexisting relationship through their mediated prime beach, meaning that beech can

activate sand (and vice versa) via the activation of beach. Similar processes could be invoked in

TOT resolution, in which the TOT word (e.g., abacus) and the mediated prime (e.g., stomach)

have a preexisting relationship through their connecting link (e.g., abdomen). That is, having a









TOT for abacus, an individual's TOT resolution may be increased by exposure to stomach,

which is the mediated prime: stomach semantically activates abdomen, which phonologically

activates abacus.

Evidence from mediated priming in both word recognition and word retrieval tasks offers

support for an interactive activation model of language production. Both discrete and interactive

activation models contend that language production involves two distinct stages of semantically

selecting and phonologically encoding a particular lemma to satisfy a communicative goal (e.g.,

Dell, 1986; Levelt, 1989). However, they disagree about the time course of these stages. Discrete

theories emphasize the successive time course of these stages, in which a particular lemma is

first selected for semantic goals and then phonologically encoded for production. Thus,

activation operates in a feed forward, top-down system. Even though multiple lemmas can be

activated for one semantic goal (e.g., near-synonyms), only the selected lemma for production

receives phonological encoding. Thus, because discrete theories do not allow for bottom-up

processing between phonemes and lemmas, these theories would predict that phonological

mediated priming cannot influence TOTs, similar to as they would predict that mediated priming

in Peterson and Savoy's (1998) task could not affect speed of naming. For example, when

stomach activates its semantic associate abdomen, bottom-up processing needs to occur for

abdomen to prime its phonological associate abacus. Conversely, interactive activation theories

postulate that these stages can co-occur via cascaded processing (Dell, 1986; MacKay, 1987;

Stemberger, 1985). That is, before a single lemma is selected to satisfy a communicative goal, all

lexical candidates are semantically and phonologically encoded. Following the "most primed

wins" principle, the lemma with the most semantic and phonological priming is selected (Burke,

et al., 1991; MacKay, 1987; MacKay & Burke, 1990). Thus, both top-down (feed forward) and









bottom-up (feedback) processing can influence lexical selection in interactive activation theories.

Within mediated priming, this interactive activation system is important as both top-down and

bottom-up processing are needed. For example, stomach relies on top-down processing to

activate its semantic associate abdomen, which relies on bottom-up processing to prime its

phonological associate abacus. Dell, Schwartz, Martin, Saffran, and Gagnon (1997) proposed

that this feed-forward and feed-back interaction between lexical candidates is functional for

fluent discourse, which can be disrupted by TOT states. However, the interaction provides more

activation to the temporarily inaccessible nodes, thereby increasing its accessibility for

production and reducing/resolving TOT states.

TOT Incidence

TOT incidence refers to the frequency of occurrence of these particular word retrieval

failures. One theoretical framework of language production, the Transmission Deficit Hypothesis

of TOTs proposes that TOT states occur because of insufficient priming to the phonological

nodes of the intended target (Burke et al., 1991; MacKay & Burke, 1990). Within this

framework, the phonological system is depicted as a hierarchical network of phonological nodes

that represent syllables, vowels, and other phonological information. These nodes are connected

by links through which node priming passes. For a word to be retrieved for production, all nodes

must be activated above threshold. However, the amount and rate of activation a node receives

depends on the strength of the links, which are weakened by frequency of use, recency of use,

and aging, as TOT states increase for low-frequency words, non-recently used words, and older

adults (Burke, MacKay, & James, 2000; Burke et al., 1991; MacKay & Burke, 1990).

To determine if strengthening the phonological connections lowers TOT incidence,

James and Burke (2000) presented participants with five primes that collectively contained all









the syllables of the upcoming target word, intermixed with five unrelated words. For example, if

the upcoming target word was abacus, participants viewed abrogate, abj ect, element, caucus, and

hibiscus, with each word containing one syllable of abacus (i.e., abrogate, abject, element,

caucus, and hibiscus). James and Burke (2000) found that relative to the control condition in

which the words were not phonologically related to the target word, presenting phonological

primes before the question decreased the incidence of TOT states. Thus, they concluded that

recent exposure to the phonological connections of a target word decreased the probability of

TOT onset, as the phonological connections have been strengthened. Experiment 1 of the current

study investigated this phenomenon via mediated priming. That is, is the activation of the

phonology of an upcoming target word via a mediated prime (a connecting link between the

target and the phonological prime) strong enough to strengthen the phonological connections,

resulting in the occurrence of fewer TOT states?

TOT Resolution

Similar to TOT incidence, strengthening of phonological connections is relevant to TOT

resolution (i.e., retrieving the intended word after having a TOT). Because TOT states are

thought to result from weakened connections between phonological nodes (Dell, 1986; MacKay,

1987), James and Burke (2000) hypothesized that phonological priming should also facilitate

TOT resolution. That is, once having a TOT state, exposure to similar phonology should

strengthen the connections of the target word, thereby increasing activation above threshold for

production. Similar to their investigation of TOT incidence, they investigated this hypothesis by

phonologically priming all syllables of a target word after presenting the question and inducing a

TOT state. For example, if in a TOT state for the target word abacus, participants viewed five

primes that collectively contained all the syllables of the target: abrogate, abj ect, element,









caucus, hibiscus. James and Burke (2000) found that relative to the control condition in which

the words were not phonologically related to the target word, presenting primes during a TOT

state increased resolution. Abrams and her colleagues (Abrams et al., 2003; White & Abrams,

2002) furthered this phenomenon, finding that priming only the first syllable of the TOT word

provided sufficient activation for TOT resolution.

The strengthening of a TOT word's phonological connections is not limited to overt

production. Abrams et al. (2003) demonstrated the effects of silent production of phonological

primes on TOT resolution. In one experiment, when in a TOT state, participants silently read a

list of words that included two phonological first-syllable primes intermixed with unrelated

words. Relative to the control condition, silent production of the primes strengthened the TOT

connections and increased TOT resolution. Together, these studies (e.g., Abrams et al., 2003;

James & Burke, 2000; White & Abrams, 2002) showed that the primes strengthened the nodes of

a target word, thereby decreasing the incidence of TOT states and increasing TOT resolution.

Experiment 2 of the current study investigated whether mediated priming has a similar

effect on TOT resolution to that of direct priming. However, even if mediated priming

strengthens the phonological connection enough for resolution, a factor that may influence

resolution is the part-of-speech of the primes and targets. Abrams and Rodriguez (2005) found

that primes of a different part-of-speech from the TOT word facilitated resolution, whereas

primes of the same part-of-speech had no effect on resolution. However, the part-of-speech

effects were correlated with frequency. For different part-of-speech primes, high-frequency

primes increased TOT resolution, whereas for same part-of-speech primes, high-frequency

primes decreased resolution. The current study used low-frequency primes (below 20) in hopes

of minimizing any impact from the prime's part of speech.









The Current Study

The current experiments examined the effects of mediated priming on TOT incidence and

resolution. In Experiment 1, participants named a near-synonymous picture that had a secondary

name to share phonology with the target word. Even if the dominant name was produced, the

secondary name was presumably activated because of spreading activation (Balota & Lorch,

1986). Participants then answered a definition-like question that corresponded to a low-

frequency word, and TOT incidence and target retrieval were measured. In Experiment 2,

participants named a near-synonymous picture that had a secondary name sharing phonology

with the target word after having a "TOT" or "Unknown" response to a definition-like question

that corresponded to a low-frequency word. If the connections are strong enough to indirectly

transmit activation to the target word, then mediated priming may be a phenomenon that occurs

in spontaneous TOT incidence and resolution. That is, being exposed to words whose activated

associates overlap in first-syllable phonology with the TOT word may decrease TOT incidence

and increase TOT resolution without individuals' awareness.

Mediated priming is hypothesized to decrease TOT incidence and increase TOT

resolution, respectively, similar to providing a phonological prime before and during a TOT

state. Because activation increasingly diminishes the further away from the activated lemma, the

activation to the target word will be reduced. Nevertheless, this amount of activation is expected

to be sufficient to strengthen the connections above threshold for TOT resolution. As seen in

Abrams et al. (2003, Exp. 3), silent production of primes (i.e., reading primes silently) increased

TOT resolution. Thus, even though participants did not overtly produce the primes, the TOT

word received enough priming activation for resolution. In the current experiments, the primes

will be similarly activated without purposeful silent production. That is, production of the










dominant picture name will activate its near-synonymous secondary name (i.e., the prime) via

spreading activation. Even though the prime is not being directly produced, its activation from

the dominant name is expected to be sufficiently activated to affect TOT incidence and

resolution.

However, when participants produce the picture's secondary name, then the name will

serve as a directly activated prime, comparable to the primes in other phonological priming

studies (e.g., Abrams et al., in press; Abrams et al., 2003; James & Burke, 2000; White &

Abrams, 2002). Similar to the mediated priming via production of the dominant name, we

predict production of the secondary name to decrease TOT incidence and increase TOT

resolution. In both instances, the TOT phonological connections are strengthened by the

activation of a prime. However, directly activated primes are expected to prime TOT incidence

and resolution even more than production of the dominant name, as production of the dominant

name requires mediated (indirect) priming.

Generally, individuals produce the dominant name of a picture, as it is most frequently

used. However, can they be more likely to produce the secondary, less frequently used name as a

function of having a TOT? A secondary investigation in Experiment 2 is the ability of a TOT

word to feed back activation to influence lemma selection. That is, does the mere activation of a

lemma in a TOT state feed back to influence the selection of the secondary (as it shares

phonology with the TOT word) rather than the dominant name in the semantic selection stage? If

the connections are strong enough to activate the secondary name as the most-primed-wins

name, then influential feedback on lemma should occur, as the participants named the picture

after TOT onset.











Cow ....*yO
PASTURE O., **** Mil


\ Calculator


Pastor


O Grass


Passage


ture


p as


Figure 1-1. Phonologically mediated priming via spreading activation. This figure demonstrates
how the activation of pasture spreads activation to related semantic and phonological associates.
Cow is a mediated prime between pasture and milk. The dashed lines represent activation sent
between words, whereas the solid lines represent activation between phonemes or words and
phonemes. The thickness of dashed lines between the nodes indicates the strength of activation
the node receives, with the thickest lines to the associates indicating the most activation.









CHAPTER 2
EXPERIMENT 1

Experiment 1 was designed to examine whether mediated priming influences TOT

incidence. If mediated priming effectively activates the phonology of the upcoming target word

via bottom-up processing, then participants should experience fewer TOT states than when not

presented with the mediated prime. A secondary aim of Experiment 1 was to assess whether

production of the dominant name influences word retrieval differently from production of the

secondary name. Production of the secondary name is a direct prime because its phonology

matches that of its target. Therefore, production of the dominant name is a mediated prime

because the phonology of the target word is indirectly activated via semantic activation of the

secondary name. Thus, because the phonology of the target word will be more strongly activated

via the direct rather than mediated prime (Dell & O'Seaghdha, 1991; 1992), we expect

participants to experience fewer TOTs after production of the secondary name, relative to

production of the dominant name.

Method

Participants

Participants included 60 young adults (50 females, 10 males), aged 18-25 (M~= 19. 13, SD

= 1.01), recruited from introductory psychology courses and given partial course credit for

participation. All reported English to be their native language.

Materials

The program was developed using Visual Basic 5.0 and conducted on a Pentium 4, 1.8

GHz PC-compatible computer. Experimental stimuli to elicit TOT states consisted of 79

definition-like questions that corresponded to a target word, with 18 borrowed from previous

TOT studies (Abrams & Rodriguez, 2005; Abrams et al., in press) and 61 novel questions









developed by the experimenter. Seventy-nine questions were selected because that was the

maximum number of appropriate target words that met the criteria to correspond to the prime

pictures. Target words were common nouns (67.09%), adj ectives (8.86%), and verbs (24.05%)

that were low in Francis and Kucera (1982; 0 to 26 per million, except one at 114 and one at 147,

M = 6.56, SD = 20.90) and between one and five syllables (M = 2.37, SD = .89).

Pilot study

Each target word was paired with an experimental, i.e., prime, picture and a control

picture. Prime pictures were nouns that had two near-synonymous names, with some obtained

from Peterson and Savoy (1998) and Snodgrass and Vanderwart (1980), and some generated by

the experimenter. Control pictures were nouns that had one name and were obtained from the

same sources as the prime pictures. All pictures were pilot tested by having 180 participants view

70 pictures (out of 350 possible) one at a time. Each participant named only 70 pictures to reduce

the chance of fatigue. Participants were asked to name the obj ect in the picture and then provide

up to three alternative names that one might also call the picture. The first name produced was

considered their dominant name, whereas the second name they produced was considered their

secondary name for the obj ect. As previously mentioned, Peterson and Savoy (1998) tabulated

the names across different responses and considered the highest frequency response the dominant

name and the less frequent response the secondary name. However, because the primes would be

designed based on the name individuals most frequently produce first to ensure they receive

mediated rather than direct priming, we tabulated names according to response position (i.e.,

first, second). Participants were given the option of providing a third response so that they did

not feel constrained when providing responses; rather, they could freely respond as the names

came to mind. From the 350 pictures pilot tested, 79 prime (near-synonymous) pictures were










selected that met the following criteria: across participants had 1) at least two appropriate,

distinctly disproportionately frequent names, 2) a name produced in the first response above

50%, and 3) a name produced in the second response above 35%. Dominant names were

produced 79.7% of the time on average, whereas subordinate names were produced on average

43.9% of the time. Furthermore, the secondary names were produced an average of 17.8% on the

first response, and the dominant names were produced an average of 14.6% on the second

response.

Stimuli for materials

For prime pictures, the first-syllable phonology of the secondary name matched that of its

target word, but not that of its corresponding dominant name. For example, the biopsy matched

in first-syllable phonology with its corresponding near-synonymous picture's secondary name

bike but not dominant name motorcycle. Of the 79 chosen prime pictures, the mean Francis and

Kucera (1982) frequency of the dominant names was 45.80 (SD = 1 12.30, range = 0 to 91, with 5

greater than 100) and of the secondary names was 3 8.80 (SD = 3086.50, range = 0 to 89, with 5

greater than 100). Additionally, the mean syllable length of the dominant names was 1.6 (SD =

.77, range = 0 to 4) and of the secondary names was 1.7 (SD = .79, range = 0 to 4). Prime

pictures and targets shared part of speech 67% of the time (both were nouns).

The control picture corresponded to only one name that had at least 95% agreement (e.g.,

pencil) and was neither phonologically nor semantically related to the target. The Francis and

Kucera (1982) frequency of the controls' names (M~= 22, SD = 44, range = 0 to 88, with 4

greater than 100) was similar to that of the prime pictures' dominant and secondary name, and

the number of syllables of the control picture matched its corresponding prime picture' s

dominant name. While control pictures were intended as a baseline against which to measure the










effects of the prime pictures, filler pictures were included to disguise the relationship between

the prime pictures and the target word. Instead of naming just the prime or control picture,

participants named two filler pictures with either the prime or control picture, resulting in 158

Eiller pictures (two pictures per target). For each target word, one filler picture corresponded to

one primary name with at least 90% agreement, e.g., anchor, and the other was a near-

synonymous obj ect corresponding to a dominant and secondary name, e.g., sheep

(dominant)/lamnb (secondary). The means for the dominant (79.7%) and secondary (43.9%) name

productions of the near-synonymous filler pictures were the same as those of the prime pictures.

The near-synonymous pictures were included as filler pictures to disguise the purpose of the

prime pictures. Additionally, each filler picture had a different first-syllable phonology and first

letter than its corresponding target, prime picture, control picture, and other filler picture.

Targets, dominant and secondary names of prime picture, and names of control pictures are

presented in Appendix A.

Design and Procedure

The experimental design was a single-factor design with Prime Condition (Prime Picture,

Control Picture) as a within-participants factor. Upon arrival to the lab, participants received

written and oral instructions that the purpose of the study was to examine TOT states. To

disguise the relation between answering the questions and naming the pictures, participants were

told that to give them a break between answering questions, they would complete a picture-

naming task for a separate study investigating names associated with certain obj ects. Before

beginning the experiment, the experimenter ensured that participants understood the definition of

a TOT, a temporary state of inaccessibility to a known word, accompanied by a feeling of










knowing, by having them explain the meaning (see Appendix B for a detailed explanation of

instructions given to participants).

To begin the study, participants viewed and named three pictures one at a time, two filler

pictures and either a prime or control picture. The prime or control picture always appeared

second, so the first and last picture that participants named were filler pictures. The computer

generated the order so that participants received half prime and half control pictures in a

randomized order. Participants then viewed a question and responded "Known", "Unknown",

"TOT", or "None of the Above" by clicking on the button next to the appropriate response (see

Figure 2-1 for an example of the order of tasks). A "Known" response was one in which

participants knew and could provide the answer, whereas an "Unknown" response was one in

which participants did not know the answer and would be unfamiliar with the word if told the

answer. A "TOT" response was defined as knowing the answer but temporarily not having access

to it, whereas a "None of the Above" response was described as knowing the answer but not

being able to retrieve it on their own.

After a "Known" response, participants indicated the answer and proceeded to the next

question. After a "TOT" response, participants were asked to indicate how many minutes would

be needed to resolve their TOT, and they then proceeded to the next question. After an

"Unknown" or "None of the Above" response, participants were asked if they could provide any

partial information (e.g., number of syllables, first letter, and so forth) about the target, and then

proceeded to the next question. This process continued until all 79 experimental questions had

been answered.

After answering the last question, participants completed the stem-completion test for the

questions for which they did not provide an answer. This test was intended to determine if










participants could retrieve the word when provided with letter cues. Each question was shown

with the first letter of its target word. If participants knew the answer from the cue of the first

letter, they provided the answer and proceeded to the next question. If they did not know the

answer, they were shown both the first and second letter to the word and given a chance to type

in the word. This process repeated until they typed in the word or were shown the maximum

number of letters of the word allowed (ranging from one to four), which was determined by the

number of letters overlapping between the prime and target's first syllable. For example, the first

two letters of the target biopsy were the maximum number of letters participants could see

because biopsy and its prime bike overlap in the first two phonemes and letters.

For the questions they did not answer on the stem-completion test, participants took a

recognition test. Each question was shown with four possible answers: the target word, a word

semantically related to the target word, a word phonologically related to the target word, and a

word semantically and phonologically unrelated to the target word. Participants were asked to

select the best answer.

Lastly, participants completed a post-experiment questionnaire that was intended to

determine whether they recognized a phonological relation between the pictures and the target

words, and if so, whether they intentionally used this association to determine the target word.

Participants were then debriefed.

Results

The responses to the post-experiment questionnaire revealed that no participants

recognized the intended relationship between the questions and prime pictures and thus did not

consciously use the relationship to influence their responses to the questions and pictures, a

finding consistent with previous TOT research (e.g., Abrams et al., in press). All analyses were










conducted by participants only (to get a mean for each participant rather than each item) because

the stimulus items were not selected randomly, a condition that has been suggested to be

inappropriate for item analyses (e.g., Cohen, 1976; Raaijmakers, Schrijnemakers, & Gremmen,

1999; Vitevitch & Sommers, 2003).

Response Types for Questions, Reported Minutes for TOT Resolution, and Picture Names

Of the initial responses to the questions, 35.7% were correct "Known" responses, 15.6%

were incorrect "Known" responses, 10.3% were correct "TOT" responses, 1.9% were incorrect

"TOT" responses, 25.3% were "Unknown" responses, and 1 1.1% were "None of the Above"

responses. A correct "Known" response was when the answer given to the general knowledge

question matched the target. A correct "TOT" response was one in which the participant

correctly retrieved the target on the stem-completion test or chose the correct answer on the

recognition test. After a "TOT" response, participants estimated that they could resolve the TOT

within 10 minutes on 94.4% of the responses (M~= 4.95 minutes, SD = 4.57, range = 1-40). Of

the responses to the prime pictures, 76.7% produced the intended dominant name, 14.3%

produced the intended secondary name, and 9.1% produced names other than the intended

dominant and secondary names.

Stem-Completion Test

Because the stem-completion test determined if participants could retrieve the word when

provided with letter cues, it examined the activation level of the target word for participants who

were and were not having a TOT. To investigate the efficacy of the stem-completion test, for

initial correct "TOT" responses, "Unknown" responses, and "None of the Above" responses, a

repeated-measures ANOVA was conducted on mean percentage of targets correctly retrieved on

the stem-completion test. The main effect of response type was significant, F(2, 88) = 16.90,










M~SE = .04, p < .001, with more correct target retrievals following "TOT" responses (M~=

66. 13%) than either "Unknown" responses (M\~= 44.68%) or "None of the Above" responses (M\~

= 44.44%), which did not differ (p > .95). Because of the nonsignificant difference between

"Unknown" and "None of the Above" responses, we combined these responses into one category

for subsequent analyses.

Mediated Priming

To test the hypothesis that mediated priming would influence TOT incidence, paired-

samples t-tests with prime condition (Prime Picture, Control Picture) were conducted separately

on the proportion of correct "Known" responses (i.e., proportion of targets correctly retrieved),

correct "TOT" responses (i.e., proportion of correct TOT states), and "Unknown"/"None of the

Above" responses. Only non-secondary name productions (i.e., dominant names as well as other

name productions that did not overlap phonologically with the target) of the prime picture were

included in the analyses, as those reflected mediated priming; the secondary name overlapped in

first-syllable phonology and therefore would be a direct phonological prime for the target if

produced. Means and standard deviations in percent are presented in Table 2-1. T-tests revealed

significant effects of prime condition. There were more "Known" responses, t(59) = 4.21, SE =

.03, p < .001, fewer "TOT" responses, t(59) = -2. 16, SE = .01, p < .04, and marginally fewer

"Unknown"/"None of the Above" responses, t(59) = -1.68, SE = .02, p < .10, following prime

pictures than control pictures. Paired-samples t-tests were then conducted on the proportion of

each response type following only dominant name productions to clarify whether including other

phonologically-unrelated responses in analysis were contributing to the above results. These t-

tests revealed similar results to those given above, with more "Known" responses, t(59) = 4.37,

SE = .02, p < .001, marginally fewer "TOT" responses, t(59) = -1.79, SE = .01, p < .08, and










fewer "Unknown"/"None of the Above" responses, t(59) = -2.07, SE = .02, p < .05, following

prime pictures.

Direct Priming

Because of its overlapping phonology with the target, production of the prime picture's

secondary name was a direct prime of the target. Therefore, direct priming on TOT incidence

was examined via paired-samples t-tests with prime condition (Prime Picture, Control Picture)

that were conducted separately on the proportion of correct "Known" responses (i.e., proportion

of targets correctly retrieved), correct "TOT" responses (i.e., proportion of correct TOT states),

and "Unknown"/"None of the Above" responses. In this analysis, only secondary name

productions of the prime picture were included in the analyses, as their first-syllable matched

that of the target. Means and standard deviations in percent are presented in Table 2-2. Unlike

the previous results on mediated priming, t-tests showed equivalent proportions of correct

"Known" responses, t < 1, correct "TOT" responses, t < 1, and "Unknown"/"None of the Above"

responses, t < 1, following production of the prime picture' s secondary name versus the control

picture.

Part of Speech

An analysis was conducted to determine if the observed mediated priming effects were

dependent on the prime and target's part of speech, i.e., whether more priming occurred when

primes and targets differed in part of speech than when they shared part of speech. Repeated-

measures ANOVAs were conducted separately on percentage of correct "Known" responses,

correct "TOT" responses, and "Unknown"/"None of the Above" responses, using Prime Part of

Speech as a factor (phonologi cally-mediated prime and target were the same part of speech,

phonologically-mediated prime and target were different parts of speech, and unrelated control










picture that was the same part of speech as the prime). Means and standard deviations in percent

from the ANOVAs are presented in Table 2-3.

For "Known" responses, there was a significant effect of Prime Part of Speech, F(2, 1 18)

= 8.34, M~SE = .02, p < .001, with more "Known" responses following same part-of-speech

primes and different part-of-speech primes than control pictures, ps < .03. There were also

marginally more "Known" responses following different part-of-speech primes than same part-

of-speech primes, p < .10. For "TOT" responses, the Prime Part of Speech effect was not

significant, F(2, 1 18) = 1.85, M~SE = .01, p > .16. Similarly, for "Unknown"/"N\one of the

Above" responses, there was no effect of Prime Part of Speech, F(2, 1 18) = 1.12, M~SE = .01, p >

.33.

Discussion

As predicted, mediated priming influenced TOT incidence, demonstrated by more

Known responses and fewer TOT responses following prime pictures relative to control pictures.

Production of the prime picture's dominant name activated the first-syllable phonology of the

upcoming target via mediated priming, allowing participants to retrieve the target more often

than when the first-syllable phonology was not activated. Even though the activation strength of

the phonology via mediated priming is thought to be weaker than that of direct priming (Dell &

O' Seaghdha, 1991, 1992), the findings suggest that the activation is strong enough to strengthen

the target's phonological connections, inducing it to be more accessible for production.

Despite Einding mediated priming, we found no evidence for direct priming on incidence

of TOT states, as production of the secondary name that overlapped phonologically with the

target did not reduce TOT states or increase Known responses. These results are contradictory to

previous findings that direct priming decreases TOT incidence (James & Burke, 2000), but the









current findings may not be a true assessment of direct priming's influence on TOT incidence

because of the low percentage of secondary name productions (14.3%), which created a large

amount of variability in participants' means and may have reduced the statistical power to detect

an effect of direct priming.

The mediated priming effects on TOT incidence occurred independent of the overlap

between the prime and target's parts of speech. Previous research has found that a direct prime's

part of speech (i.e., whether it shares the same or has a different part of speech than the target)

affects TOT resolution differently (Abrams & Rodriguez, 2005; Abrams et al., in press);

different part-of-speech primes increase TOT resolution, whereas same part-of-speech primes

have no effect. The current study also found effects due to the mediated prime's part of speech on

TOT incidence, but both same- and different part-of-speech primes facilitated target retrieval by

increasing "Known" responses, relative to control pictures, and different part-of-speech primes

helped marginally more than same part-of-speech primes. While the different part-of-speech

findings were predicted by the TDH, the facilitatory effect of same part-of-speech primes was

not. The present study involved TOT incidence, whereas the previous studies on part of speech

have only investigated their effects on TOT resolution, so perhaps TOT resolution is more

sensitive to grammatical class than TOT incidence.

In sum, these findings demonstrated another process by which TOT states can be

prevented from occurring in the first place. In addition to direct exposure to a phonologically-

similar prime (James & Burke, 2000), TOT incidence may also be lowered via mediated

priming, as the activation sent to related associates is strong enough to activate the node above

threshold for production. Experiment 2 investigated mediated priming effects on TOT resolution.

While TOT incidence and TOT resolution both involve strengthening the phonological









connections to the selected word for production, decreasing incidence requires strengthening

prior to lemma selection, whereas increasing resolution requires strengthening after lemma

selection. Determining if mediated priming can affect TOT incidence and resolution similar,

similar to findings with direct priming (James & Burke, 2000), will provide further

understanding of the processes involved in strengthening the phonological connections to both

prevent and resolve TOT states.










Table 2-1. Initial responses (in %) following production of the prime pictures' non-secondary
name and control pictures' name in Experiment 1

Prime Condition
N Prime Picture Control Picture
Initial Response Mean SD Mean SD

Known 60 42.21 19.23 31.66 15.35
TOT 60 8.75 7.55 11.73 9.03
Unknown/None of the Above 60 34.65 17.25 37.63 16.23










Table 2-2. Initial responses (in %) following production of the prime pictures' secondary name
and control pictures' name in Experiment 1

Prime Condition
N Prime Picture Control Picture
Initial Response Mean SD Mean SD

Known 60 33.07 25.66 31.33 15.41
TOT 60 11.07 13.83 11.80 8.96
Unknown/None of the Above 60 38.80 25.13 37.65 16.15










Table 2-3. Initial responses (in %) following prime pictures that shared the part of speech (POS)
with the target, prime pictures that differed in part of speech from the target, and control pictures
in Experiment 1


Prime Condition
N Same POS Prime Different POS Prime


Control


Picture
Initial Response


Mean


Mean


SD Mean SD


Known
TOT
Unknown/None of the Above


60 37.99
60 10.00
60 35.27


21.49
9.98
17.83


41.75 20.20
8.96 8.59
35.07 18.11


31.33
11.80
37.65


15.41
8.96
16.15




















What noun means the
removal of a sample of
tissue, which is then
examined under a
microscope to check for
cancer cells?


Anchor






Motorcycle
(Bike)







Sheep
(Lamb)


SBiopsy
(if Known)


Respond: Known,
TOT, Unknown, or
None of the Above


n


View Question


Name Pictures






Figure 2-1. Order of tasks for Experiment 1. Participants named three pictures (two
filler pictures and either a prime or control picture) and then answered a general
knowledge question. The question was associated with a target word that was
phonologically related to the secondary name of the prime picture. The target names
of the pictures are below the corresponding picture. The secondary name of the
prime picture is in parentheses.









CHAPTER 3
EXPERIMENT 2

Experiment 2 was designed to examine whether mediated priming influences TOT

resolution, i.e., retrieval of targets following TOT states. Because mediated priming effectively

influenced TOT incidence in Experiment 1, i.e., because the activation from the dominant name

(e.g., motorcycle) to the secondary name (e.g., bike) was strong enough to activate the phonology

for the target word to decrease TOT incidence, we expected the same strength of activation to

facilitate TOT resolution. In addition to mediated priming, we also investigated differences in

resolution following direct priming (production of the secondary name). Because direct priming

has been shown to successfully facilitate TOT resolution (e.g., Abrams et al., 2003; Abrams et

al., in press), we predicted greater resolution following direct priming relative to no priming, as

the target receives more activation directly from a prime rather than indirectly (Dell &

O' Seaghdha, 1991, 1992).

A secondary examination in this experiment was the ability of a TOT word to feed back

activation to influence lemma selection; that is, can having a TOT induce participants to produce

the secondary name of the picture, the phonologically-related name to the TOT word, more often

than they would otherwise? We expected more secondary name productions when participants

were in a TOT state (i.e., the target activates the secondary name's phonology via bottom-up

processing) relative to when they did not know the answer (i.e., the target is not activated, so

bottom-up processing cannot occur to activate secondary name's phonology).

Method

Participants

Participants included 60 young adults (41 females, 19 males), aged 18-25 (M~= 18.92, SD

= 1.13), recruited from introductory psychology courses and given partial course credit for










participation. All considered English to be their native language and did not participate in

Experiment 1.

Materials

The same materials in Experiment 1 were used in Experiment 2.

Design and Procedure

The experimental design and procedure for Experiment 1 was used for Experiment 2,

except for the order of the tasks. That is, participants named a picture after their response to the

questions (see Figure 3-1 for example). Thus, participants viewed a question associated with a

low-frequency word and responded "Known", "Unknown", "TOT", or "None of the Above".

After a "Known" response, participants typed their response, named a control and two filler

pictures, and proceeded to the next question. After a "TOT", "Unknown", or "None of the

Above" response, participants named two filler pictures and either a prime or control picture, and

then attempted to answer the same question before proceeding to the next question.

Results

Similar to Experiment 1, no participants indicated on the post-experiment questionnaire

that they recognized the intended relationship between the questions and prime pictures or

consciously used the relationship to influence their responses to the questions and pictures.

Response Types for Questions, Reported Minutes for TOT Resolution, and Picture Names

Of the initial responses to the questions, 36.8% were correct "Known" responses, 15.1%

were incorrect "Known" responses, 11.1% were correct "TOT" responses, 1.9% were incorrect

"TOT" responses, 24.9% were "Unknown" responses, and 10.3% were "None of the Above"

responses. After a "TOT" response, participants estimated that they could resolve the TOT within

10 minutes on 91.8% of the responses (M~= 5.43 minutes, SD = 6.57, range = 1-43). Of the










responses to the prime pictures, 70.7% were the intended dominant name, 20.9% were the

intended secondary name, and 8.4% were names other than the intended dominant and secondary

names.

Stem-Completion Test

Similar to Experiment 1, the stem-completion test examined the activation level of the

target word for participants who were and were not having a TOT. For initial correct "TOT"

responses, "Unknown" responses, and "None of the Above" responses, a repeated-measures

ANOVA was conducted on mean percentage of correctly retrieved targets on the stem-

completion test. Fourteen participants were excluded from this analysis because they did not

have to take the stem-completion test at least once following each response type. The main effect

of response type was significant, F(2, 90) = 15.84, M~SE = .02, p < .001, with more correct target

retrievals following "TOT" responses (66.21%) than "Unknown" (44.41%) and "None of the

Above" (46. 18%) responses, which did not differ from each other (p > .66). Because of the

nonsignifieant difference between "Unknown" and "None of the Above" responses, these

responses were combined in subsequent analyses.

Mediated Priming

A 2 (Response Type: correct "TOT", "Unknown"/"N\one of the Above") x 2 (Prime

Condition: Prime Picture, Control Picture) repeated-measures ANOVA was conducted on the

proportion of targets correctly retrieved following the picture naming task. A correct "TOT"

response was one in which the participant responded "TOT" when they first saw the question,

and either 1) responded "Known" and provided the correct answer when viewing the question a

second time; 2) provided the correct answer on the stem-completion test; 3) correctly identified

the answer on the recognition test. This analysis included only non-secondary name productions










(dominant and other phonologi cally-unrelated responses), as those productions measure

mediated priming, whereas secondary name productions assessed direct priming. Four

participants were excluded from this analysis because they did not provide the non-secondary

name of the prime picture at least in once in all of the conditions. The means and standard

deviations from this analysis converted into percent are shown in Table 3-1.

The main effect of prime condition was significant, F(1, 55) = 22. 19, M~SE = .12, p <

.001, showing more correct target retrievals following prime pictures than control pictures. The

main effect of response type was also significant, F(1, 55) = 50.03, M~SE = .06, p < .001, with

greater resolution after correct "TOT" responses than after "Unknown"/"None of the Above"

responses. These main effects were moderated by a significant Response Type x Prime

Condition interaction, F(1, 55) = 5.39, M~SE = .04, p < .03, shown in Figure 3-2. Follow-up tests

on the Response Type x Prime Condition interaction revealed that significant priming occurred

after initial correct "TOT" responses, p < .001, with more correct target retrievals following

prime pictures than control pictures. There was also significant priming after initial

"Unknown"/"None of the Above" responses, p < .001, but to a lesser extent. There was greater

target retrieval following correct "TOT" responses than "Unknown"/"None of the Above"

responses for both prime conditions, with a larger difference for the prime pictures, ps < .001.

To assess whether including other phonologically-unrelated responses in the above

analyses contributed to the findings, a 2 (Response Type: correct "TOT", "Unknown"/"None of

the Above") x 2 (Prime Condition: Prime Picture, Control Picture) repeated-measures ANOVA

was conducted on target retrieval following dominant name productions only. Seven participants

were excluded from the analysis because they did not have at least one intended dominant name










production in all of the conditions. The analysis revealed identical results to the previous

analy si s.

Direct Priming

A 2 (Response Type: correct "TOT", "Unknown"/"N\one of the Above") x 2 (Prime

Condition: Secondary Name Production of Prime Picture, Control Picture) repeated-measures

ANOVA was conducted on the proportion of targets correctly retrieved following the picture

naming task. Thirty-three participants were excluded from this analysis because they did not

provide the secondary name of the prime picture at least once following both response types. The

means and standard deviations from this analysis converted into percent are shown in Table 3-2.

The main effect of prime condition was significant, F(1, 26) = 24.58, MSE = .23, p <

.001, showing more correct target retrievals following secondary name productions of prime

pictures than control pictures. The main effect of response type was also significant, F(1, 26) =

16.09, MSE = .07, p < .001, with greater resolution after correct "TOT" responses than after

"Unknown"/"N\one of the Above" responses. These main effects were moderated by a marginally

significant Response Type x Prime Condition interaction, F(1, 26) = 2.98, MSE = .04, p < .10.

Follow-up tests on the Response Type x Prime Condition interaction revealed that significant

priming occurred after initial correct "TOT" responses, p < .001, with more correct target

retrievals following secondary name productions of prime pictures than control pictures. There

was also significant priming after initial "Unknown"/"N\one of the Above" responses, p < .001,

but to a lesser extent. There was greater target retrieval following correct "TOT" responses than

"Unknown"/"N\one of the Above" responses for both prime conditions, with a larger difference

for the prime pictures, ps < .001.









To compare whether there was greater priming of target retrieval following secondary

versus non-secondary name production, a 2 (Response Type: correct "TOT", "Unknown"/"None

of the Above") x 3 (Prime Condition: Secondary Name Production of Prime Picture, Non-

Secondary Name Production of Prime Picture, Control Picture) was conducted on the proportion

of targets correctly retrieved. Thirty-three participants were excluded from this analysis because

they did not have at least one correct "TOT" response and one "Unknown"/"None of the Above"

response in each of the prime conditions. The main effect of response type was significant, F(1,

26) = 26.79, M~SE = .07, p < .001, with more correct target retrievals following correct "TOT"

responses (44.6%) than "Unknown"/"None of the Above" (23.8%) responses. Additionally, the

main effect of prime condition was significant, F(2, 52) = 20.57, M~SE = .14, p < .001, where

both secondary and non-secondary name production led to greater target retrieval than the

control picture (8.2%), ps < .001, indicating significant priming for both types of names.

However, more correct target retrievals occurred following secondary name production (52.4%)

than non-secondary name production (42.1%), p < .005. The Response Type x Prime Condition

interaction was not significant, F(2, 52) = 1.62, M~SE = .04, p > .20.

Part of Speech

Potential part of speech effects on mediated priming were examined via a 2 (Response

Type: Correct "TOT", "Unknown"/"None of the Above") x 3 (Prime Part of Speech:

phonologically-mediated prime and target were the same part of speech, phonologically-

mediated prime and target were different parts of speech, and unrelated control picture that was

the same part of speech as the prime) repeated-measures ANOVA on target retrieval. Twenty

participants were excluded from the analysis because they did not have at least one correct

"TOT" response and "Unknown"/"None of the Above" response in each of the prime part-of-










speech categories. The means and standard deviations from this analysis converted into percent

are shown in Table 3-3. The main effect of response type was significant, F(1, 39) = 53.58, M\SE

=.07, p < .001, showing more correct target retrievals following correct "TOT" than

"Unknown"/"N\one of the Above" responses. The main effect of prime part of speech was also

significant, F(2, 78) = 16.01, M~SE = .13, p < .001, with greater resolution after same and

different part-of-speech primes than control pictures, p < .001, but no difference in target

retrieval between same and different part-of-speech primes, p > .15. The Response Type x Prime

Part of Speech interaction, F(2, 78) = 1.60, M~SE = .05, p > .20, was not significant.

Feedback

To examine if the TOT word can feed back to influence picture name production, a

paired-samples t-test was conducted on percentage of secondary name productions following

correct "TOT" responses and "Unknown"/"N\one of the Above" responses. There was no

difference between secondary name productions after correct "TOT" responses (M~= 19.8%) and

"Unknown"/"N\one of the Above" responses (M~= 20.2%), t < 1.

The feedback effect was also examined by comparing percentage of secondary name

productions in Experiment 2 with those in Experiment 1 (where the picture was named first

before viewing the questions, so there was no opportunity for feedback from the unretrieved

target). Independent-samples t-tests were conducted on percentage of secondary name

productions between the two experiments, separately for correct "TOT" responses and

"Unknown"/"N\one of the Above" responses. Following "TOT" responses, there were more

secondary name productions in Experiment 2 (19.8%) than in Experiment 1 (14.5%), t(1 17) = -

1.74, SE = .03, p < .001. A similar pattern emerged for secondary name productions following









"Unknown/None of the Above" responses in Experiment 2 (20.2%) than Experiment 1(14.5%),

t(118) =-2.49, SE = .02, p <.001.

Discussion

Mediated priming increased TOT resolution by strengthening the needed phonological

connections for target retrieval, evidenced by greater target retrieval following production of the

prime pictures' dominant name relative to production of control pictures. Even though the first-

syllable phonology of the TOT word was only indirectly activated (i.e., production of the

dominant name activated its semantic associate, the secondary name, which spread phonological

activation to the TOT word), the activation strength was strong enough to strengthen the

phonological connections for TOT resolution. Similar to TOT incidence, this mediated priming

effect was independent of part of speech, as both same and different part-of-speech primes

facilitated TOT resolution. The hypothesis that direct priming would influence TOT resolution

was also supported, as production of prime pictures' secondary name increased TOT resolution

relative to control pictures. In addition to replicating previous findings on the facilitation of TOT

resolution via direct phonological priming (e.g., Abrams et al., 2003; Abrams et al., in press;

James & Burke, 2000), the current study extends these Eindings by showing that direct priming is

more effective for resolving TOTs than mediated priming.

In addition to TOT states, direct and mediated priming also unexpectedly facilitated

target retrieval after "Unknown"/"None of the Above" responses, a contradictory finding from

previous TOT research that has shown virtually no target retrieval following "Unknown"

responses (Abrams & Rodriguez, 2005; Abrams et al., 2006; James & Burke, 2000). One

methodological difference between previous studies and the current study is that we included

participants' estimates of their TOT resolution time. Participants in both Experiment 1 and 2









estimated that they could resolve the maj ority of their TOT states (over 90% of them) within 10

minutes. One possibility is that asking participants to report the time to resolve their TOT may

have caused them to be more conservative about classifying a state a TOT state. Therefore, they

classified only strong TOT states, ones in which they could resolve within 10 minutes, as TOT

states; weaker TOT states with a less sense of imminent retrieval may have instead been reported

as "Unknown"/"N\one of the Above" responses. Benefiting from priming, the TOT states

included in the Unknown/None of the Above category would cause the category to show priming

effects. The results from mediated priming's influence on TOT incidence supports this

explanation, as fewer "Unknown"/"None of the Above" responses occurred following prime than

control pictures. If participants were in a true "unknown" state, there should be no change in

these responses following priming. Furthermore, the results from the stem-completion test

support this misclassification explanation, as targets were retrieved at a fairly high rate following

"Unknown" (44. 14%) and "None of the Above" (46. 18%) responses, although not as high

following "TOT" responses (66.21%); if participants had responded "Unknown" only when the

target was truly unknown, the retrieval rate on the stem-completion test would be expected to be

closer to 0%. Additionally, despite its priming effect, the mediated and direct priming effects for

"Unknown"/"None of the Above" responses were smaller than those for "TOT" responses, which

supports the misclassification theory. Because the activation strength of both the TOT state and

the prime determines if the TOT is activated above threshold for production, weak TOT states

will need more activation strength from primes to get resolved than strong TOT states.

Therefore, because the TOT states in "Unknown"/"None of the Above" and "TOT" responses

received the same primes, i.e., received the same activation, then the priming effect difference









between "TOT" and "Unknown"/"N\one of the Above" responses must be attributable to the

activation strength of the TOT states.

The prediction that secondary name productions would be more prevalent when

participants were in a TOT state than when in an "unknown" state, as the activated TOT word

would feed back to activate the phonology of the secondary name enough for production, was

not supported. Following correct "TOT" responses, participants were no more likely to produce

the secondary name than after "Unknown"/"N\one of the Above" responses, suggesting that the

bottom-up processing from TOT states was not strong enough to influence lemma selection. An

alternative explanation is that the lack of secondary name productions follow "TOT" and

"Unknown"/ "None of the Above" responses could be due to the picture naming task; that is, the

secondary name could always be activated for retrieval, an activation that may be greater than

what is experienced in the TOT state. Therefore, regardless of whether in a TOT state,

participants may produce the secondary name merely because it is always activated with the

dominant name. However, the comparison of secondary name productions in Experiment 1 and

Experiment 2 does demonstrate the occurrence of a feedback effect. Because participants in

Experiment 1 named the pictures before viewing the question, the feedback effect was not

possible; thus, Experiment 1 served as a control group to which secondary name productions in

Experiment 2 were compared. The feedback effect was possible in Experiment 2 because

participants named the picture after viewing the question.

For both TOT and "Unknown"/"N\one of the Above" responses, participants were more

likely to produce the secondary name in Experiment 2 than Experiment 1. In the cases in which

participants produced the secondary name in Experiment 2, the activation to the secondary name

was strong enough to exceed that of the dominant name, the most frequently produced name, to









become the most-primed-wins node and get produced. Because this effect also occurred when

participants said they were in an "unknown" state, we cannot decisively conclude that the

activated TOT lemma is responsible for the effect.

However, if participants misclassified their weaker TOTs as "unknown" states as

suggested earlier, then the partial activation of TOT states that would occur in both TOT and

Unknown/None of Above categories may be strong enough to transmit enough phonological

activation to related lemmas for production. If this partial activation of TOTs explanation is

correct, then the feedback effect should have been smaller for "Unknown"/"None of the Above"

responses than "TOT" responses, similar to the smaller effects of direct and mediated priming on

"Unknown"/"None of the Above" relative to "TOT" responses. The finding that the feedback

effect was similar for both response types suggests that the phonological activation strength of

weak TOTs was sufficient to influence lemma selection similar to that of strong TOTs. Thus,

misclassified TOTs, i.e., weak TOTs, may influence TOT resolution and feedback differently:

More phonological activation may be required for a word to emerge from a TOT state (TOT

resolution) than for a word to exceed the activation of the more frequently-produced dominant

name (feedback). Thus, the activation strength of TOTs seems more important in TOT resolution

than feedback. In TOT resolution, the word became a TOT state because of weak phonological

connections (Burke et al., 1991; MacKay, 1987). For the word to be retrieved from its TOT state,

the phonological activation had to exceed threshold. Because of their greater activation strength,

strong TOTs needed less activation from the primes than weak TOTs; thus, TOT resolution

occurred more for strong than weak TOTs. In the feedback instances, the common, familiar

secondary name was different from the TOT word because their phonological connections were

strong enough to not become a TOT state. Therefore, the secondary name was activated enough









when participants viewed the picture to only need a small amount of additional phonological

activation from the TOT lemma to become the most-primed-wins node over the dominant name,

meaning that the phonological activation from the weak TOT may be sufficient for the name to

be produced (strong TOTs just provide extra activation that is not needed). In turn, TOTs may

influence the lexical system sometimes better than they can be influenced.

In sum, similar to its influence on TOT incidence, mediated priming facilitated TOT

resolution by indirectly activating the first-syllable phonology of the TOT word above threshold

for production. These findings demonstrate that even though the activation strength sent to target

words via mediated priming is weaker than that of direct priming (Dell & O' Seaghdha 1991,

1992), it is sufficient to influence TOT states. Thus, comparable to direct priming (e.g., Abrams

et al., in press; James & Burke, 2000), mediated priming is an effective priming method by

which spontaneous TOTs may be resolved. Collectively, these studies demonstrate that mediated

priming can benefit TOT states by decreasing incidence and increasing resolution.










Table 3-1. Target retrieval (in %) following production of the prime pictures' non-secondary
name and control pictures' name in Experiment 2


Prime Condition


N Prime Picture
Mean SD


Control Picture
Mean


Initial Response


TOT
Unknown/None of the Above


46.63 40.39
18.36 27.87


18.64
2.37


28.76
5.10










Table 3-2. Target retrieval (in %) following production of the prime pictures' secondary name
and control pictures' name in Experiment 2


Prime Condition


Prime Picture
Mean S


Control Picture
Mean SD


Initial Response


TOT
Unknown/None of the Above


65.93
38.91


45.60
44.07


14.82
1.57


24.02
3.72










Table 3-3. Target retrieval (in %) following prime pictures that shared the part of speech (POS)
with the target, prime pictures that differed in part of speech from the target, and control pictures
in Experiment 2


Prime Condition
Different POS Prime


N Same POS Prime


Control

Mean SD


Picture
Initial Response


Mean

55.38
26.50


Mean

50.05
21.21


TOT 40
Unknown/None of the Above 40


40.71
36.12


45.43
34.43


19.85
1.88


29.24
4.11



















What noun refers
to a system or
condition of
society in which
the means of
production are
owned and
controlled by the
state?


View Question
(if Response was
TOT, Unknown,
or None of the
Above)


Tree






Couch
(Sofa)






Necklace
(Pearl s)


Socialism
(if Known)


Respond: Known,
TOT, Unknown, or
None of the Above


View Question


Name Pictures







Figure 3-1. Order of tasks for Experiment 2. Participants viewed a question, provided their
response, and then named three pictures (two filler pictures and either a prime or
control picture). The question was associated with a target word that was
phonologically related to the secondary name of the prime picture. The target
names of the pictures are below the corresponding picture. The secondary name
of the prime picture is in parentheses.












60

c, 50

40 i I Prime Picture

30 i I Control Picture

o t 20


10
Correct TOT Unknown/None of the Above
Res ponse Type
Figure 3-2. Percent correct target retrieval following prime and control pictures for "TOT" and
"Unknown"/"None of the Above" responses in Experiment 2.









CHAPTER 4
GENERAL DISCUSSION

Analogous to Peterson and Savoy's (1998) demonstration of effective mediated priming

in word recognition, the current study extended this phenomenon to word retrieval in relation to

TOT states. These results coincide with previous TOT evidence that support interactive

activation models of language production (Dell, 1986; MacKay, 1987; Stemberger, 1985),

thereby supporting Dell et al.'s (1997) hypothesis that interactive activation is beneficial for

fluent discourse. Because the phonological activation can occur before the selection of a lemma

(i.e., the most activated lemma to be selected as the most-primed-wins node), discourse is more

fluent as the nodes can be prepared for production via priming. In the case of TOT incidence, the

lemma needed for production can receive phonological activation via mediated priming before

lemma selection, thereby decreasing the incidence of TOT states. If the lemma does not receive

sufficient activation prior to selection, it can nevertheless receive additional priming after

selection, even if the individual is exposed to the phonology only indirectly. Therefore, mediated

priming in TOT states furthers the notion that bottom-up processing is an important part of the

lexical system, a process that is not predicted by discrete theories (Levelt, 1989).

In addition to interactive activation models, the findings that mediated priming influences

TOT states supports the recency hypothesis of the TDH (Burke, et al., 2000; Burke et al., 1991;

MacKay & Burke, 1990). Because the target's phonological connections were recently

strengthened via the indirect activation of the phonologically-related secondary name, mediated

priming decreased TOT incidence and increased TOT resolution, similar to the process thought

to underlie direct priming from phonologically-related words (James & Burke, 2000). In

conclusion, mediated priming has parallel effects on TOT incidence and resolution, as the same

processes to strengthen phonological connections help to reduce as well as resolve TOT states.










Replicating previous TOT research (Abrams & Rodriguez, 2005; Abrams et al., in press;

Abrams et al., 2003), the current study found that direct phonological priming of the TOT word' s

first syllable facilitated TOT resolution. A new contribution to the understanding of TOT states

is that direct priming is more effective in resolution than mediated priming, as was predicted by

Dell and O' Seaghdha (1991, 1992). Naming the picture's secondary name provided more

activation to the phonological connections, thereby providing a greater chance of TOT

resolution, because the phonology was a direct recipient of the activation. In mediated priming,

the phonology is an indirect recipient that received a proportion of the original activation; thus,

the probability of the TOT state receiving enough activation for production decreased.

Although not explicitly manipulated, the current study found part-of-speech to affect

mediated priming of TOT incidence and resolution similarly, where both different and same part-

of-speech primes decreased TOT incidence (measured by an increase in "Known" responses) and

increased TOT resolution. However, while different part-of-speech primes decreased TOT

incidence more than same part-of-speech primes, there was no difference in TOT resolution

following same and different part-of-speech primes. These findings contrast with Abrams and

Rodriguez (2005) and Abrams et al. (in press), who found that primes of a different part-of-

speech from the TOT word facilitated resolution, whereas primes of the same part-of-speech

resolution did not increase target retrieval. The similar, facilitatory effects of same and different

part-of-speech primes on TOT states in the present study may be attributable to the frequency of

the secondary names of the primes (i.e., frequency of picture name productions given as the

second response on the pilot picture naming task; M~= 38.80), which was lower than those in

Abrams et al. (2006; M~= 63.01). Frequency of same part-of-speech primes has been shown to

negatively correlate with TOT resolution (Abrams & Rodriguez, 2005). Therefore, the primes










may have been too low in frequency to serve as potential competitors with the target for

retrieval, which is necessary to delay retrieval as explained by the "most primed wins" principle

of the Node Structure Theory (MacKay, 1987). The activation of a node in the same syntactic

class as the target word should not help retrieval because the activated prime would be the most-

primed-wins node for that class. However, not being activated as much as higher-frequency

primes, low-frequency primes are activated enough to provide phonological activation to the

target word but not compete to be the most-primed-wins node. Furthermore, as Dell and

O' Seaghdha (1991, 1992) explained, mediated primes are activated less and, thus, send less

activation to the target than direct primes because they receive only a proportion of the original

activation from their activated associate. Because they do not have enough activation to be

activated above threshold for production, mediated primes may not be as strong of a competitor

in their grammatical class to be the most-primed-wins node as direct primes. Therefore, a target

word in the same grammatical class can receive phonological activation from the mediated prime

but not have to compete with the partially-activated prime.

The feedback effect was demonstrated through TOT states, as the partially-activated

words influenced lemma selection. Additionally, if participants misclassified their weak TOT

states in the Unknown/None of the Above category, the results reveal that weak TOT states may

have the same influence on the lexical systems as strong TOT states, which the further support

the interactive activation theories, as bottom-up processing was required for the activated

phonology to influence lemma selection. Additionally, these findings further our understanding

of TOT states and their role in the lexical system. However, further investigations are necessary

to clarify the relationship between the TOT state and the feedback effect. Instead of measuring

the feedback effect by secondary name productions of near-synonymous pictures, future research









could investigate picture naming times following TOT and unknown states. If TOT states can

uniquely feed back to prime lemmas, then participants should name the pictures faster following

TOT states than unknown states. To specifically to determine if weak TOT states can influence

lemma selection similarly to strong TOT states, future research could similarly investigate

picture naming times but distinguish between weak and strong TOTs based on feeling-of-

knowing judgments. If the TOT state does influence lemma selection, the finding will not only

demonstrate the activation strength of TOT states but also their role in the lexical selection (i.e.,

similar to the way in which they receive phonological priming, they can provide phonological

priming to related lemmas). Additionally, the feedback effect could be used understand the

duration of TOT states: Can the feedback effect occur for TOT states that have been unresolved

for 20 minutes similar to those that have been unresolved for 5 minutes? Theoretically, the

sooner the TOT onset to lemma selection of another word, the most activation the TOT word

and, thus, greater influence it will have on the lexical system.

Whereas this experiment investigated the TDH's recency hypothesis, future research

might investigate the TDH's aging hypothesis by examining age-related influences of mediated

priming on TOT states. Because aging weakens phonological connections (e.g., MacKay &

Burke, 1990), node priming diminishes with age. In mediated priming, the TOT state being

activated above threshold for production relies on a function of the activation of distant

associates and the strength of the phonological connections. Thus, if one of these variables is

weak, as the latter one is in aging, then mediated priming cannot activate the TOT word above

threshold, which may be seen in older adults, particularly old-old adults. These results may have

several informative outcomes. First, they would coincide with previous evidence demonstrating

age-related changes in direct phonological priming of TOT resolution (Abrams et al., in press;









White & Abrams, 2002) and offer support for the TDH's aging hypothesis. Second, the results

would help explain why older adults experience more TOT states than younger adults: Younger

adults benefit from both direct and mediated priming to decrease TOT incidence, whereas older

adults may only benefit from direct priming. Third, they would demonstrate that mediated

priming may not be an effective method for investigating cognitive processes in older adults.

The previously discussed possibility of the misclassification of weak TOT states is an

evident limitation to the current study. Although the results demonstrate that mediated priming is

an effective priming strategy, we cannot decisively conclude that the feedback effect is due to

TOT states. Additionally, while we attempted to include clear near-synonymous pictures, i.e.,

ones that have a clear dominant and secondary name that mean the same thing, we had to resort

to using some that were more of semantic associates than near-synonyms (e.g., world and globe)

to increase the stimuli size. However, near-synonyms and semantic associates may differ

lexically, which could affect their role in mediated priming. For example, nzailnzan and postman

are clear near-synonyms, they can be appropriately used interchangeably and have the exact

same definitions. Therefore, when either the definition or picture ofnzailmnzanpostnzan is

presented, theoretically the node representing the definition should send equal activation to both

nzailnzan and postman. Receiving equal activation as it near-synonym, the node that had the most

activation prior to the definition or picture being presented, e.g., nzailnzan, is the one that is

selected for production. However, being the secondary name, the other node, e.g., postman, can

still influence the lexical system, as it is partially activated. Conversely, the node representing

the definition or picture of a globe should only send activation to globe, as it is the only term that

accurately fits the definition or describes the picture. Because they are semantic associates, globe

can then send activation to world; however, because world did not receive its activation from the










original source, i.e., the picture of the globe, globe will have more activation strength than world.

Therefore, postman, the secondary name in the near-synonym example, should theoretically have

more activation to influence lemmas than world, the secondary name in the semantic associate

example. Understanding the potential difference of near-synonyms and semantic associates will

further the theoretical understanding of lemma interaction in the lexical system.

The influence of mediated priming of TOTs is an intriguing phenomenon, as individuals

are seemingly unaware of phonological similarity of the primes and the TOT words. These

findings offer another means of understanding spontaneous TOT incidence and resolution. In

addition to James and Burke's (2000) explanation that spontaneous TOT resolution may occur

when the individual hears or sees a word phonologically related to the TOT word, spontaneous

TOT resolution may occur via mediated priming. That is, the individual may hear or see a word

that semantically activates a word phonologically related to the TOT word. When actively

attempting to resolve the TOT state, individuals tend to think of same part of speech alternatives

(Burke et al., 1991), which may actually hinder the resolution incidence (Abrams et al., in press,

Abrams & Rodriguez, 2005). Thus, a beneficial method to induce resolution may merely be to

not actively pursue TOT resolution, as it may be achieved via mediated priming. In everyday

life, this phenomenon may simply mean that the individual disregards the TOT state, allowing

for inadvertent exposure to the TOT word's first syllable phonology, whether via direct or

mediated priming, may induce "spontaneous" TOT resolution. Given the high number of

spontaneous resolutions in everyday life, it seems likely that mediated priming is responsible:

Mediated priming provides individuals with more opportunities for resolution than would be

expected if only direct priming influences TOT states.










APPENDIX A
OF PRIME PICTURES, SECONDARY NAMES OF
AND UNRELATED CONTROL PICTURES


TARGETS, DOMINANT NAMES
PRIME PICTURES,


Dominant Name
of Prime Picture


Secondary Name
of Prime Picture


Unrelated Control
Picture


Target


acquit
aorta
badminton
bailiff
ballad
biopsy
bistro
boa
bogey
bookworm
bootee
b otani st
bunion
cadaver
calamine
caviar
cherub
chives
chronic
chute
compost
covet
dagger
di charge
dungeon
eligible
fluoride
garland
genealogy
gingko
gloat
hemorrhage
hostel
jester
justice
kaleidoscope
kosher
liable


fish tank
gorilla
scale
bikini
purse
motorcycle
mixer
canoe
raft
hedges
stereo
chest
rabbit
tape
money
hat
recliner
baby
alligator
boot
policeman
cabinet
ballerina
plates
weights
moose
tile
trashcan
pants
present
world
chicken
pig
airplane
pitcher
mug
jacket
lamp


aquarium
ape
balance
bathing suit
bag
bike
beaters
boat
boat
bu shes
boombox
box
bunny
cassette
cash
cap
chair
child
crocodile
shoe
cop
cupboard
dancer
dishes
dumbbells
elk
floor
garbage can
jeans
gift
globe
hen
hog
jet
jug
cup
coat
light


compass
tomato
dice
lawn mower
heart
helicopter
elbow
pretzel
disk
rainbow
cheerleader
grapes
eyebrow
door
football
star
typewriter
tiger
rhinoceros
cat
swing
trophy
trampoline
train
calendar
sun
ball
penguin
well
handcuffs
house
puzzle
rain
mushroom
angel
lock
pumpkin
slide










liquidate
lotus
lumberj ack
lynch
mallet
metronome
mi sogyni st
mount
mutiny
nocturnal
nomad
papyrus
pasteurize
patriarchal
patronize
pawn
photosynthesis
pivot
potpourri
ritual
salvage
satire
saturate
sherbet
socialism
spay
spectrum
spritz
stock
stoic
streamline
strobe
stutter
taxidermy
ten or
toga
torpedo
turban
typhoon
veto
warden


branch
bread
suitcase
chain
rat
ruler
glove
lips
donkey
book
binder
nightgown
pot
bucket
glue
toilet
pictures
needle
mailman
bow
flip-flops
sub
bags
blouse
couch
rocket
glasses
runner
traffic light
rock
road
carriage
ab domen
license plate
teepee
frog
turtle
record player
wheel
car
cane


limb
loaf
luggage
links
mouse
measuring stick
mitt
mouth
mule
novel
notebook
paj amas
pan
pail
paste
potty
photographs
pin
postman
ribbon
sandal s
sandwich
sacks
shirt
sofa
spaceship
spectacles
sprinter
stop light
stone
street
stroller
stomach
tag
tent
toad
tortoi se
turntable
tire
vehicle
walking stick


kite
tank
arrow
hinge
pipe
spider
fan
queen
cactus
hamburger
mermaid
flower
snail
camera
wrench
skateboard
balloon
hammer
horseshoe
squirrel
castle
clock
cake
tree
key
basket
feather
peacock
paperclip
cross
microscope
yo-yo
violin
skunk
band-aid
sewing machine
pifiata
fire hydrant
harp
leaf
roller coaster









APPENDIX B
EXPERIMENT INSTRUCTIONS

Upon arrival to the lab, participants read instructions indicating that the purpose of the

experiment was to study tip-of-the-tongue (TOT) states. Participants read the following

description to understand a TOT state:

A TOT occurs when you are unable to retrieve a word that you are certain you know.
When experiencing a TOT, you may know the word's definition, you know how you
want to use it in a sentence, and sometimes you can say what letter it starts with or what it
sounds like. A TOT state is different from just feeling like you "should" know the word;
in a TOT state, you may have the feeling that the word is just beyond your reach, or you
may feel a sense of frustration because of your inability to recall it. If you are uncertain
what a TOT state is, please ask the experimenter to describe it more fully now.

After reading the instructions, all participants indicated that they understood a TOT state, but the

experimenter reiterated the definition provided in the instructions.

The instructions stated that we would study TOT states by having participants complete

two tasks. The first task would be answering definition-like, general knowledge questions that

ask participants to produce the word that best fits the definition. After viewing the question, they

are to respond in one of four ways: a) "Known" if they know and can provide the answer; b)

"Unknown" if they do not know the answer and would not know it if someone told them; c)

"TOT" if they know the word, but just cannot seem to access it at that moment, though possibly

feeling as though it is "right there"; d) "None of the Above" if they know the word but are not in

a TOT state for it; that is, they do not have that feeling that it is "right there", but if someone told

them the word, they would know it. Participants read that they should give their response

promptly after reading the question, and that the questions were intended to be difficult so to not

worry if they do not know all of the answers.

The instructions described the second task as an unrelated task that involves naming

pictures of obj ects to give them a break from answering the questions. Participants read that we










were collecting data for another study on names of common obj ects, so they would see various

pictures. They were instructed to type the first name that comes to mind for the obj ect when they

see a picture. Additionally, they read that because we are interested in their first impression,

there is no right or wrong answer, to feel free to give multiple words to name the picture if that is

what comes to mind, and to not worry about the spelling. Lastly, they were informed that all

pictures were nouns.

After participants read the instructions, the experimenter ensured that participants

understood the procedure by reiterating the nature of the two tasks and the response options to

the questions. When the experimenter felt assured that participants understood the procedure,

participants were allowed to begin the experiment.










LIST OF REFERENCES


Abrams, L., & Rodriguez, E. L. (2005). Syntactic class influences phonological priming of tip-
of-the-tongue resolution. Psychonomic Bulletin & Review, 12, 1018-1023.

Abrams, L., Trunk, D. L., & Merrill, L. A. (in press). Why a superman cannot help a tsunami:
Activation of grammatical class influences resolution of young and older adults' tip-of-
the-tongue states. Psychology and Aging.

Abrams, L., White, K.K., & Eitel, S.L. (2003). Isolating phonological components that increase
tip-of-the-tongue resolution. Memory & Cognition, 31, 1153-1162.

Balota, D.A., & Lorch, R.F. (1986). Depth of automatic spreading activation: Mediated priming
effects in pronunciation but not in lexical decision. Journal ofExperimentalPsychology:
Lea-l~rning. Memory, and' Cognition, 12, 336-345.

Brown, R., & McNeill, D. (1966). The "tip of tongue" phenomenon. Journal of Verbal Learning
& Verbal Behavior, 5, 325-337.

Burke, D.M., MacKay, D. G., & James, L.E. (2000). In T.J. Perfect & E.A. Maylor (Eds.),
Models of cognitive aging (pp. 204-237). New York: Oxford University Press.

Burke, D.M., MacKay, D., Worthley, J.S., & Wade, E. (1991). On the tip of the tongue: What
causes word finding failures in young and older adults? Journal of2~emory & Language,
30, 542-579.

Cohen, J. D. (1976). Random means random. Journal of Verbal Learning & Verbal Behavior,
15, 261-262.

Cutting, J.C., & Ferreira, V.S. (1999). Semantic and phonological information flow in the
production lexicon. Journal ofExperimental Psychology: Lea-l~rning. Memory, and'
Cognition, 25, 318-344.

Dell, G.S. (1986). A spreading-activation model of retrieval in sentence production.
Psychological Review, 93, 283-321.

Dell, G.S., & O' Seaghdha, P.G. (1991). Mediated and convergent lexical priming in language
production: A comment on Levelt et al. (1991). Psychological Review, 98, 604-614.

Dell, G.S., & O'Seaghdha, P.G. (1992). Stages of lexical access in language production.
Cognition, 42, 287-3 14.

Dell, G.S., Schwartz, M.F., Martin, N., Saffran, E.M., & Gagnon, D.M. (1997). Lexical access in
aphasic and nonaphasic speakers. Psychological Review, 104, 801-838.










Francis, W.N., & Kucera, H. (1982). Frequency Analysis of English Usage: Lexicon and
Grananar. Boston, MA: Houghton-Mifflin Company.

James, L.E., & Burke, D.M. (2000). Phonological priming effects on word retrieval and tip-of-
the-tongue experiences in young and older adults. Journal ofExperintentalPsychology:
Lea-l~rnin. M~entory, and Cognition, 26, 1378-1391.

Jeschniak, J.D., & Schriefers, H. (1998). Discrete serial versus cascaded processing in lexical
access in speech production: Further evidence from the coactivation of near-synonyms.
Journal ofExperintental Psychology: Lea-l~rning. Mentory, and Cognition, 24, 1256-1274.

Levelt, W.J.M. (1989). Speaking: From intention to articulation. Cambridge, MA: MIT Press.

MacKay, D. (1987). The organization of perception and action. A theory for language and other
cognitive skills. New York: Springer.

MacKay, D. & Burke, D.M. (1990). Cognition and aging: A theory of new learning and the use
of old connections. In T. Hess (Ed.). Aging and cognition: Knowledge organization and
utilization (pp. 213-263). Amsterdam: North Holland.

Peterson, R.R., & Savoy, P. (1998). Lexical selection and phonological encoding during
language production: Evidence for cascaded processing. Journal ofExperintental
Psychology: Lea-l~rning. Mentory, and Cognition, 24, 539-557.

Raaijmakers, J. G. W., Schrijnemakers, J. M. C., & Gremmen, F. (1999). How to deal with "the
language-as-fixed-effect fallacy": Common misconceptions and alternative solutions.
Journal of~entory & Language, 41, 416-426.

Snodgrass, J.G., & Vanderwart, M. (1980). A standardized set of 260 pictures: Norms for name
agreement, image agreement, familiarity, and visual complexity. Journal ofExperintental
Psychology: Human Learning and Mentory, 6, 174-215.

Stemberger, J.P. (1985). An interactive activation model of language production. In A.W. Ellis
(Ed.), Progress in the psychology of language (Vol. 1, pp. 143-186. Hillsdale, NJ:
Erlbaum.

Vitevitch, M. S., & Sommers, M. S. (2003). The facilitative influence of phonological similarity
and neighborhood frequency in speech production in younger and older adults. M~enory
and Cognition, 31, 491-504.

White, K.K., & Abrams, L. (2002). Does priming specific syllables during tip-of-the-tongue
states facilitate word retrieval in older adults? Psychology & Aging. 1 7, 226-23 5.

White, K.K., & Abrams, L. (2004). Phonologically mediated priming of preexisting and new
associations in young and older adults. Journal ofExperintental Psychology: Lea-l~rning.
M~entoy, and Cognition, 30, 645-655.









BIOGRAPHICAL SKETCH

Lisa Merrill was born in Monroe, North Carolina. After graduating from Forest Hills

High School, she attended Furman University where she earned her Bachelor of Arts in

psychology. Later that same year, she began conducting research with Dr. Lise Abrams at

University of Florida and will earn her Master of Science in psychology in 2007.





PAGE 1

INFLUENCES OF MEDIATED PRIMING ON TIP-OF-THE-TONGUE INCIDENCE AND RESOLUTION By LISA A. MERRILL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007 1

PAGE 2

2007 Lisa A. Merrill 2

PAGE 3

To my parents 3

PAGE 4

ACKNOWLEDGEMENTS This project would not have been possible wi thout the support of many people. First, I thank my advisor, Dr. Lise Abrams, for not onl y her continuous help in refining my ideas, developing stimuli, and revising the paper but also her incessant encouragement and motivation that kept me enthused and optimistic about my work. I thank my supervisory committee, Dr. Wind Cowles and Dr. Ira Fischler, for their time and effort to help me successfully and professionally complete this project, and th eir thoughtful encouragement through their kind words. I thank Kati Brown and Sara Fernandez fo r their generous help and humor during stimuli development. Finally, I thank Will, Julie, David, Matt, and my parents for always providing me with unconditional love and support in my decisions and pursuits. 4

PAGE 5

TABLE OF CONTENTS ACKNOWLEDGEMENTS. LIST OF TABLES...6 LIST OF FIGURES. ABSTRACT....8 CHAPTER 1 INTRODUCTION. 2 EXPERIMENT 1...21 Method...21 Results....26 Discussion..30 3 EXPERIMENT 2... Method...37 Results....38 Discussion.. 4 GENERAL DISCUSSION4 APPENDIX A Targets, Dominant Names of Prime Pictur es, Secondary Names of Prime Pictures, and Unrelated Control Pictures..........60 B Experiment Instructions.....................62 REFERENCE LIST...64 BIOGRAPHICAL SKETCH.66 5

PAGE 6

LIST OF TABLES Table Page 2-1 Initial responses (in %) following production of the prime pictures non-secondary name and contro l pictures' name in Experiment 1...33 2-2 Initial responses (in %) following production of the prime pictures secondary name and control pictures' name in Experiment 1..... ....4 2-3 Initial responses (in %) following following prime pictures that shared the part of speech (POS) with the target, prime pictures that differed in part of speech from the target, and control pictures in Experiment 1................5 3-1 Target retrieval (in %) following production of the prime pictures non-secondary na me and control pictures' name in Experiment 2..9 3-2 Target retrieval (in %) following produc tion of the prime pi ctures secondary name and control pictures' name in Experiment 2.......50 3-3 Target retrieval (in %) following prime pi ctures that shared the part of speech (POS) with the target, pr ime pictures that differed in part of speech from the target, and control pictures in Experiment 2....51 6

PAGE 7

LIST OF FIGURES Figure Page 1-1 Phonologically mediated prim ing via spreading activation...20 2-1 Order of tasks for Experiment 1.....36 3-1 Order of tasks for Experiment 2.....52 3-2 Percent correct target retrieval follo wing prime and control pictures for "TOT" and "Unknown"/"None of the Above" responses in Experiment 2....3 7

PAGE 8

Abstract of Thesis Presente d to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science INFLUENCES OF MEDIATED PRIMING ON TIP-OF-THE-TONGUE INCIDENCE AND RESOLUTION By Lisa A. Merrill December 2007 Chair: Lise Abrams Major: Psychology Previous research indicate s that tip-of-the-tongue (TOT) states, characterized by the temporary inability to retrieve a known wor d, can be reduced as well as resolved via phonological priming, i.e., presenta tion of prime words phonologically related to the TOT target word. The current study investigated the influenc es of mediated phonological priming, in which the prime and target were indirectly related, on the incidence and reso lution of TOT states. Participants answered genera l knowledge questions designed to induce a TOT state and named pictures in which one was either a prime or unr elated picture. The prime was a near-synonymous picture that corresponded to a dominant (e.g., motorcycle) and secondary (e.g., bike ) name, where the dominant name referred to the name typically produced upon seeing the picture. The secondary name was phonologically related to the TOT word by sh aring its first syllable (e.g., biopsy). In Experiment 1, participants named a near-synonymous picture before answering a general knowledge question, and the results showed that producing the dominant name of prime pictures increased target retrie val and decreased TOT states for subsequently presented targets, relative to naming unrelated pict ures. In Experiment 2, particip ants named a near-synonymous picture after answering the general knowledge questions, and for targets that were in TOT states, 8

PAGE 9

naming the prime pictures' dominant name increased resolution of those TOT states compared to unrelated pictures. Because production of the dom inant name led to decreased TOT incidence and increased TOT resolution than did an unrelat ed name, these results demonstrate that the phonologically-related secondary name (without bei ng produced) primed the ta rget for retrieval. In addition to providing another method of primi ng TOT states for future research, exposure to the TOT word's first-syllable phonology via mediated priming may be a process involved in the spontaneous resolution of TO T states in everyday life. 9

PAGE 10

CHAPTER 1 INTRODUCTION Tip-of-the-tongue (TOT) states, characterized by the temporary inability to retrieve a known word (e.g., Brown & McNeill, 1966), are comm on disruptions to fluent speech that are not always immediately resolved (e.g., Burke, MacKay, Worthley, & Wade, 1991). The cause of TOT states is thought to result from weakened phonological c onnections, which makes the TOT word unable to receive sufficient activati on for production (e.g., MacKay, 1987; MacKay & Burke, 1990). Support for this view comes from pr evious research, in which presenting primes that are phonologically related to the target word can facilitate TOT resolution (e.g., seeing abdomen primes the TOT word abacus ; Abrams & Rodriguez, 2005; Abrams, Trunk, & Merrill, in press; Abrams, White, & Eitel, 2003; Jame s & Burke, 2000; White & Abrams, 2002). This method is thought to mimic the spontaneous resolu tion of TOTs in everyday life, as exposure (i.e., via seeing or hearing) to a TOT words phonological connections induces the word to be more accessible but without a person's conscious knowledge (James & Burke, 2000). The current experiments examined TOT incidenc e and resolution via phonologicallymediated priming, which may be another means of spontaneous TOT re solution. That is, instead of direct exposure to the first-syllable phonology of the TOT word, TOT incidence and resolution may also occur because of mediated exposure to the first-syllable phonology of the TOT word, without actually encountering that phonology directly. Mediated Priming Mediated priming involves primes and target s that are indirectly related by a connecting link (e.g., pasture and milk are indirectly relate d by their connecting link cow ) and is beneficial for investigating factors of sp reading activation. Spreading ac tivation models contend that activation of a lemma (i.e., the semantically and syntactically specified re presentation, which is 10

PAGE 11

the word concept and corresponding properties) spreads to distant, indirectly-rela ted and close, directly-related associates, where relationship can be defined at multiple levels, such as meaning (semantic) or sound (phonology; Balota & Lorch, 1986). For example, as seen in Figure 1-1, because milk receives activation through pasture s directly-related semantic associate cow, milk is an indirectly-related semantic associate of pasture. In addition to semantic associates, mediated priming also occurs with phonological associates. For example, calculator is a phonological, indirectly-r elated associate of pasture through the mediated prime cow These latter types of mediated primes were used in the present experiments to influence TOT incidence and resolution. Even though indirectly-related semantic and phonological associat es are activated via spreading activation, Dell and OSeaghdha (1991 1992) proposed that these lemmas receive only a proportion of the original activation. In Figure 1, the connector lines between the lemmas demonstrate this notion. For exam ple, the fully activated prime pasture cannot send its full amount of activation to its associate cow as it sends some of its ac tivation to other associates (e.g., passage grass etc.). Of its received activation, the associate cow sends a share of its activation to its associate milk Even though indirectly-related associates receive only a proportion of the original activation, they can nevertheless influence activity in the lexical system (e.g., prime other words for activation). T hus, when considering mediated primings role in word retrieval, even though the prime is indirec tly related to the target word, the target word may still receive enough activation to decrease TOT incidence and increase TOT resolution. Previous research has used mediated prim ing to investigate th e processes underlying word recognition, word retrieval, phonological processing, and lexical activation (e.g., Cutting & Ferreira, 1999; Jeschniak & Schriefers, 1998; Peterson & Savoy, 1998; White & Abrams, 2004). 11

PAGE 12

For example, Peterson and Savoy (1998) demonstrat ed the effects of medi ated priming on word recognition1 using near-synonyms, which are two or mo re words that correspond to the same target. The most frequently used word is referre d to as the dominant name, and the other, less frequently used word(s) are the secondary names. For example, a long upholstered article of furniture to seat more than one person can be appropriately referred to as either couch (dominant) or sofa (secondary). Peterson and Savoy (1998) first determined the pictures dominant and secondary names by ha ving participants name a series of pictures for which there was more than one appropriate name. A picture was considered near-synony mous if participants provided two basic-level names in at least 75% of the names given, and the frequency of the names was distinctly disproportionate. The most frequently used name, averaged across all responses, was the dominant name, which was pro duced a mean of 84% (range of 67% 98%), and the less frequently used name was considered the secondary name. To examine the effects of mediated priming on word recognition, participants viewed a near-synonymous picture and an experimental wo rd that was either phon ologically related (e.g., soda) or unrelated (e.g., horse) to the secondary name (e.g., sofa ), and they produced the experimental word aloud. Participants named the phonologically-related word faster than the phonologically-unrelated word, demonstrating that word recognition was facilitated via mediated priming. Viewing a near-synonymous picture sp read activation to both the corresponding dominant and secondary names, which sim ilarly activated their semanticallyand phonologically-related lemmas. Being phonologically related to the secondary name, the ___________________________________ 1Peterson and Savoy (1998) described their ta sk as representing both word recognition and production. However, participants read aloud words presented on the screen; they did not retrieve the word from memory, as the word was provi ded for them. Therefore, I refer to the task in terms of word recognition on ly to differentiate it from tasks that require self-initiated production. 12

PAGE 13

experimental word received enough priming to be activated and produced more quickly, relative to the phonologically-unrelated word Another alternative explanation may be that instead of the dominant name spreading activa tion to the secondary name, priming may have occurred only for participants who silently named the picture with the secondary name, ther eby directly activating the phonology of the experimental word. However, these results of mediated priming have been replicated in similar word r ecognition studies using Dutch words (Jeschniak & Schriefers, 1998) and homonyms (Cutting & Ferreira, 1999), suggesting that mediated priming did occur in this study. Furthermore, White and Abrams (2004) de monstrated mediated priming in word retrieval by using an associate word-stem completion task. Participants completed word-stems in one of three conditions: 1) a semantic condition, in which a word was paired with the first letter of a specific semantic associate (e.g., beach-s___ with sand as the target); 2) a phonological condition, in which a word was paired with the fi rst letter of its homophones semantic associate (e.g., beech-s___ ), and 3) an unrelated condition, in whic h an orthographically similar word to the homophones was paired with the first le tter used in the other conditions (e. g., batch-s ___). White and Abrams (2004) found that relative to the unrelated condition, participants produced the semantic associate more ofte n in the phonological condition (i.e., sand was produced more often following beech than batch ). In this example, beech activates its phonol ogical associate, beach which in turn activates its ow n semantic associates, such as sand. Thus, beech and sand have a preexisting relationship through their mediated prime beach, meaning that beech can activate sand (and vice versa) vi a the activation of beach Similar processes could be invoked in TOT resolution, in which the TOT word (e.g., abacus ) and the mediated prime (e.g., stomach) have a preexisting relationship through their connecting link (e.g., abdomen ). That is, having a 13

PAGE 14

TOT for abacus, an individual's TOT resolution may be increased by exposure to stomach, which is the mediated prime: stomach semantically activates abdomen, which phonologically activates abacus Evidence from mediated priming in both word recognition and word re trieval tasks offers support for an interactive activa tion model of language production. Both discrete and interactive activation models contend that language production involves two dis tinct stages of semantically selecting and phonologically encodi ng a particular lemma to satisfy a communicative goal (e.g., Dell, 1986; Levelt, 1989). However, they disagree a bout the time course of these stages. Discrete theories emphasize the successive time course of these stages, in which a particular lemma is first selected for semantic goals and th en phonologically encoded for production. Thus, activation operates in a feed forward, top-down system. Even though multiple lemmas can be activated for one semantic goal (e.g., near-synonyms), only the selected lemma for production receives phonological encoding. Thus, because discrete theories do not allow for bottom-up processing between phonemes and lemmas, thes e theories would predict that phonological mediated priming cannot influence TOTs, similar to as they would predict that mediated priming in Peterson and Savoys (1998) ta sk could not affect speed of naming. For example, when stomach activates its semantic associate abdomen bottom-up processing needs to occur for abdomen to prime its phonological associate abacus Conversely, interactiv e activation theories postulate that these stages can co-occur via cascaded proces sing (Dell, 1986; MacKay, 1987; Stemberger, 1985). That is, before a single lemma is selected to satisfy a communicative goal, all lexical candidates are semantically and phonologically encoded. Following the "most primed wins" principle, the lemma with the most semantic and phonologica l priming is selected (Burke, et al., 1991; MacKay, 1987; MacKay & Burke, 1990). Thus, both top-down (feed forward) and 14

PAGE 15

bottom-up (feedback) processing can influence lexical selection in interactive activation theories. Within mediated priming, this in teractive activation system is important as both top-down and bottom-up processing are needed. For example, stomach relies on top-down processing to activate its semantic associate abdomen which relies on bottom-up processing to prime its phonological associate abacus Dell, Schwartz, Martin, Sa ffran, and Gagnon (1997) proposed that this feed-forward and feed-back interactio n between lexical candida tes is functional for fluent discourse, which can be disrupted by TOT states. However, the interaction provides more activation to the temporarily inaccessible nodes, thereby incr easing its accessibility for production and reducing/resolving TOT states. TOT Incidence TOT incidence refers to the frequency of o ccurrence of these partic ular word retrieval failures. One theoretical framework of language production, the Transmi ssion Deficit Hypothesis of TOTs proposes that TOT states occur because of insufficient priming to the phonological nodes of the intended target (Burke et al., 1991; MacKay & Burke, 1990). Within this framework, the phonological system is depicted as a hierarchical network of phonological nodes that represent syllables, vowels, and other phonological information. These nodes are connected by links through which node priming passes. For a word to be retrieved for production, all nodes must be activated above threshold. However, the amount and rate of ac tivation a node receives depends on the strength of the links, which are w eakened by frequency of use, recency of use, and aging, as TOT states increase for low-freque ncy words, non-recently used words, and older adults (Burke, MacKay, & James, 2000; Bu rke et al., 1991; MacKay & Burke, 1990). To determine if strength ening the phonological connections lowers TOT incidence, James and Burke (2000) presented participants with five primes that collectively contained all 15

PAGE 16

the syllables of the upcoming target word, intermix ed with five unrelated words. For example, if the upcoming target word was abacus participants viewed abrogate, abject, element, caucus, and hibiscus, with each word containing one syllable of abacus (i.e., abrogate, abject, el e ment, cau cus, and hibis cus ). James and Burke (2000) found that relative to the c ontrol condition in which the words were not phonologically relate d to the target word, presenting phonological primes before the question decr eased the incidence of TOT stat es. Thus, they concluded that recent exposure to the phonological connections of a target word decreased the probability of TOT onset, as the phonological connections have been strengthened. Experiment 1 of the current study investigated this phenomenon via mediated priming. That is, is the activation of the phonology of an upcoming target word via a medi ated prime (a connecting link between the target and the phonological prime) strong enough to strengthen the phonological connections, resulting in the occurrence of fewer TOT states? TOT Resolution Similar to TOT incidence, strengthening of phonological connections is relevant to TOT resolution (i.e., retrieving the intended word after having a TOT). Because TOT states are thought to result from weakened connections between phonological nodes (Dell, 1986; MacKay, 1987), James and Burke (2000) hypothesized that phonological priming should also facilitate TOT resolution. That is, once having a TOT st ate, exposure to si milar phonology should strengthen the connections of the target word, thereby increasing activat ion above threshold for production. Similar to their investig ation of TOT incidence, they investigated this hypothesis by phonologically priming all syllables of a target word after presen ting the question and inducing a TOT state. For example, if in a TOT state for the target word abacus participants viewed five primes that collectively contained all the syllables of the target: ab rogate, abject, el e ment, 16

PAGE 17

cau cus, hibiscus. James and Burke (2000) found that rela tive to the control condition in which the words were not phonologically related to the target word, presenting primes during a TOT state increased resolution. Abrams and her colleagues (Abrams et al., 2003; White & Abrams, 2002) furthered this phenomenon, finding that prim ing only the first syllable of the TOT word provided sufficient activation for TOT resolution. The strengthening of a TOT words phonologica l connections is not limited to overt production. Abrams et al. (2003) de monstrated the effects of s ilent production of phonological primes on TOT resolution. In one experiment, when in a TOT state, participants silently read a list of words that included tw o phonological first-syllable primes intermixed with unrelated words. Relative to the control condition, silent production of the primes strengthened the TOT connections and increased TOT resolution. Toge ther, these studies (e.g., Abrams et al., 2003; James & Burke, 2000; White & Abrams, 2002) showed that the primes strengthened the nodes of a target word, thereby decreasing the incidence of TOT states and incr easing TOT resolution. Experiment 2 of the current study investigated whether mediated priming has a similar effect on TOT resolution to that of direct priming. However, even if mediated priming strengthens the phonological connection enough for resolution, a factor that may influence resolution is the part-of-speech of the primes and targets. Abrams a nd Rodriguez (2005) found that primes of a different part-of-speech fr om the TOT word facilitated resolution, whereas primes of the same part-of-speech had no effect on resolution. However, the part-of-speech effects were correlated with frequency. For di fferent part-of-speech primes, high-frequency primes increased TOT resolution, whereas for same part-of-speech primes, high-frequency primes decreased resolution. The current study us ed low-frequency primes (below 20) in hopes of minimizing any impact from the prime's part of speech. 17

PAGE 18

The Current Study The current experiments examined the effect s of mediated priming on TOT incidence and resolution. In Experiment 1, par ticipants named a near-synonymous picture that had a secondary name to share phonology with the target word. Even if the dominant name was produced, the secondary name was presumably activated beca use of spreading activ ation (Balota & Lorch, 1986). Participants then answered a definitio n-like question that corresponded to a lowfrequency word, and TOT incidence and target retrieval were measured. In Experiment 2, participants named a near-synonymous picture that had a secondary name sharing phonology with the target word after having a "TOT" or "Unknown" response to a definition-like question that corresponded to a low-frequency word. If the connections are strong enough to indirectly transmit activation to the target word, then medi ated priming may be a phenomenon that occurs in spontaneous TOT incidence a nd resolution. That is, being expos ed to words whose activated associates overlap in first-syllable phonology with the TOT word may decrease TOT incidence and increase TOT resolution withou t individuals awareness. Mediated priming is hypothesized to decr ease TOT incidence and increase TOT resolution, respectively, similar to providing a phonological prime before and during a TOT state. Because activation increasingly diminishes the further away from th e activated lemma, the activation to the target word will be reduced. Neve rtheless, this amount of activation is expected to be sufficient to strengthen the connections above threshold for TOT resolution. As seen in Abrams et al. (2003, Exp. 3), silent production of primes (i.e., read ing primes silently) increased TOT resolution. Thus, even though participants did not overtly produce the primes, the TOT word received enough priming activation for resolu tion. In the current experiments, the primes will be similarly activated without purposeful silent production. That is, production of the 18

PAGE 19

dominant picture name will activate its near-syn onymous secondary name (i.e., the prime) via spreading activation. Even though the prime is not being directly produced, its activation from the dominant name is expected to be suffici ently activated to affect TOT incidence and resolution. However, when participants produce the picture's secondary name, then the name will serve as a directly activated prime, comparable to the prim es in other phonological priming studies (e.g., Abrams et al., in press; Abrams et al., 2003; James & Burke, 2000; White & Abrams, 2002). Similar to the mediated prim ing via production of the dominant name, we predict production of the secondary name to decrease TOT incidence and increase TOT resolution. In both instances, the TOT phonologi cal connections are strengthened by the activation of a prime. However, directly activated primes are expected to prime TOT incidence and resolution even more than production of the dominant name, as prod uction of the dominant name requires mediated (indirect) priming. Generally, individuals produce the dominant name of a picture, as it is most frequently used. However, can they be more likely to produ ce the secondary, less frequently used name as a function of having a TOT? A secondary investigation in Experime nt 2 is the ability of a TOT word to feed back activation to influence lemma selection. That is, does the mere activation of a lemma in a TOT state feed back to influence the selection of the secondary (as it shares phonology with the TOT word) rather than the domina nt name in the semantic selection stage? If the connections are strong enough to activate th e secondary name as the most-primed-wins name, then influential feedback on lemma should occur, as the participants named the picture after TOT onset. 19

PAGE 20

20 PASTURE ture Passage pas Pasto Cow Milk r Calculator Grass as ure t p Figure 1-1. Phonologically mediated prim ing via spreading activation. This figure demonstrates how the activation of pasture spreads activation to related semantic and phonological associates. Cow is a mediated prime between pasture and milk The dashed lines represent activation sent between words, whereas the solid lines represent activation between phonemes or words and phonemes. The thickness of dashed lines between the nodes indicates the strength of activation the node receives, with the thickest line s to the associates indicating the most activation.

PAGE 21

CHAPTER 2 EXPERIMENT 1 Experiment 1 was designed to examine whether mediated priming influences TOT incidence. If mediated priming effectively ac tivates the phonology of the upcoming target word via bottom-up processing, then participants should experience fewer TOT states than when not presented with the mediated prime. A secondary aim of Experiment 1 was to assess whether production of the dominant name influences word retrieval differently from production of the secondary name. Production of the secondary na me is a direct prime because its phonology matches that of its target. Therefore, producti on of the dominant name is a mediated prime because the phonology of the target word is indir ectly activated via sema ntic activation of the secondary name. Thus, because the phonology of the ta rget word will be more strongly activated via the direct rather than mediated pr ime (Dell & OSeaghdha, 1991; 1992), we expect participants to experience fewer TOTs after production of the seconda ry name, relative to production of the dominant name. Method Participants Participants included 60 young adults ( 50 females, 10 males), aged 18-25 ( M = 19.13, SD = 1.01), recruited from introductory psychology c ourses and given partial course credit for participation. All reported Eng lish to be their native language. Materials The program was developed using Visual Basic 5.0 and conducted on a Pentium 4, 1.8 GHz PC-compatible computer. Experimental s timuli to elicit TOT states consisted of 79 definition-like questions that corresponded to a target word, with 18 borrowed from previous TOT studies (Abrams & Rodriguez, 2005; Abrams et al., in press) and 61 novel questions 21

PAGE 22

developed by the experimenter. Seventy-nine qu estions were selected because that was the maximum number of appropriate target words th at met the criteria to correspond to the prime pictures. Target words were common nouns ( 67.09%), adjectives (8.86%), and verbs (24.05%) that were low in Francis and Kucera (1982; 0 to 26 per million, except one at 114 and one at 147, M = 6.56, SD = 20.90) and between one and five syllables (M = 2.37, SD = .89). Pilot study Each target word was paired with an expe rimental, i.e., prime, picture and a control picture. Prime pictures were nouns that had two near-synonymous names, with some obtained from Peterson and Savoy (1998) and Snodgrass and Vanderwart (1980), and some generated by the experimenter. Control pictur es were nouns that had one name and were obtained from the same sources as the prime pictures. All pictures were pilot tested by having 180 participants view 70 pictures (out of 350 possible) one at a time. Each participant named only 70 pictures to reduce the chance of fatigue. Participants were asked to name the object in the picture and then provide up to three alternative names that one might also call the pictur e. The first name produced was considered their dominant name, whereas the s econd name they produced was considered their secondary name for the object. As previously mentioned, Peterson and Savoy (1998) tabulated the names across different responses and consider ed the highest frequency response the dominant name and the less frequent response the secondary name. However, because the primes would be designed based on the name individuals most fre quently produce first to ensure they receive mediated rather than direct priming, we tabulated names according to response position (i.e., first, second). Participants were given the option of providing a third response so that they did not feel constrained when provi ding responses; rather, they c ould freely respond as the names came to mind. From the 350 pictures pilot te sted, 79 prime (near-synonymous) pictures were 22

PAGE 23

selected that met the following criteria: across participants had 1) at least two appropriate, distinctly disproportionately frequent names, 2) a name produced in the first response above 50%, and 3) a name produced in the second re sponse above 35%. Dominant names were produced 79.7% of the time on average, whereas subordinate names were produced on average 43.9% of the time. Furthermore, the secondary names were produced an average of 17.8% on the first response, and the dominant names were produced an average of 14.6% on the second response. Stimuli for materials For prime pictures, the first-syllable phonology of the secondary name matched that of its target word, but not that of its corresponding dominant name. For example, the biopsy matched in first-syllable phonology with its correspondin g near-synonymous picture's secondary name bike but not dominant name motorcycle. Of the 79 chosen prime pictures, the mean Francis and Kucera (1982) frequency of the dominant names was 45.80 ( SD = 112.30, range = 0 to 91, with 5 greater than 100) and of th e secondary names was 38.80 (SD = 3086.50, range = 0 to 89, with 5 greater than 100). Additionall y, the mean syllable length of the dominant names was 1.6 ( SD = .77, range = 0 to 4) and of the secondary names was 1.7 ( SD = .79, range = 0 to 4). Prime pictures and targets shared part of sp eech 67% of the time (both were nouns). The control picture corresponded to only one na me that had at least 95% agreement (e.g., pencil) and was neither phonologically nor semantic ally related to the target. The Francis and Kucera (1982) frequency of the controls names (M = 22, SD = 44, range = 0 to 88, with 4 greater than 100) was similar to that of the prime pictures do minant and secondary name, and the number of syllables of the control pict ure matched its corresponding prime pictures dominant name. While control pict ures were intended as a baseline against which to measure the 23

PAGE 24

effects of the prime pictures, filler pictures we re included to disguise the relationship between the prime pictures and the target word. Instead of naming just the prime or control picture, participants named two f iller pictures with eith er the prime or control picture, resulting in 158 filler pictures (two pictures per target). For each target word, one fille r picture corresponded to one primary name with at least 90% agreement, e.g., anchor and the other was a nearsynonymous object corresponding to a dominant and secondary name, e.g., sheep (dominant)/ lamb (secondary). The means for the domina nt (79.7%) and secondary (43.9%) name productions of the near-synonymous filler pictures we re the same as those of the prime pictures. The near-synonymous pictures were included as filler pictures to disguise the purpose of the prime pictures. Additionally, each filler picture had a different first-syllable phonology and first letter than its corresponding target, prime pictur e, control picture, and other filler picture. Targets, dominant and secondary names of prim e picture, and names of control pictures are presented in Appendix A. Design and Procedure The experimental design was a single-factor design with Pr ime Condition (Prime Picture, Control Picture) as a within-participants factor. Upon arrival to the lab, participants received written and oral instructions that the purpos e of the study was to examine TOT states. To disguise the relation between answ ering the questions and naming the pictures, participants were told that to give them a break between answ ering questions, they would complete a picturenaming task for a separate study investigating na mes associated with certain objects. Before beginning the experiment, the experimenter ensure d that participants understood the definition of a TOT, a temporary state of inaccessibility to a known word, accompanied by a feeling of 24

PAGE 25

knowing, by having them explain the meaning (s ee Appendix B for a detailed explanation of instructions given to participants). To begin the study, participants viewed and na med three pictures one at a time, two filler pictures and either a prime or control picture. The prime or c ontrol picture always appeared second, so the first and last picture that particip ants named were filler pictures. The computer generated the order so that participants receiv ed half prime and half control pictures in a randomized order. Participants then view ed a question and res ponded "Known", "Unknown", "TOT", or "None of the Above" by clicking on th e button next to the a ppropriate response (see Figure 2-1 for an example of the order of tasks). A "Known" response was one in which participants knew and could provide the answ er, whereas an "Unknown" response was one in which participants did not know the answer and would be unfamiliar with the word if told the answer. A "TOT" response was defined as knowing the answer but temporarily not having access to it, whereas a "None of the Above" response was described as knowing the answer but not being able to retrieve it on their own. After a "Known" response, participants indicat ed the answer and proceeded to the next question. After a "TOT" response, participants we re asked to indicate how many minutes would be needed to resolve their TOT, and they th en proceeded to the next question. After an "Unknown" or "None of the Above" response, participants were asked if they could provide any partial information (e.g., number of syllables, first letter, and so forth) about the target, and then proceeded to the next question. This process co ntinued until all 79 experimental questions had been answered. After answering the last questi on, participants completed the stem-completion test for the questions for which they did not provide an answ er. This test was intended to determine if 25

PAGE 26

participants could retrieve the word when pr ovided with letter cues. Each question was shown with the first letter of its target word. If participants knew the answer from the cue of the first letter, they provided the answer and proceeded to the next question. If they did not know the answer, they were shown both the first and second letter to the word and given a chance to type in the word. This process repeated until they typed in the word or were shown the maximum number of letters of the word allowed (ranging from one to four), which was determined by the number of letters overlapping betw een the prime and target's first syllable. For example, the first two letters of the target biopsy were the maximum number of letters participants could see because biopsy and its prime bike overlap in the first two phonemes and letters. For the questions they did not answer on th e stem-completion test participants took a recognition test. Each question wa s shown with four possible answ ers: the target word, a word semantically related to the target word, a word phonologically related to the target word, and a word semantically and phonologically unrelated to the target word. Participants were asked to select the best answer. Lastly, participants completed a post-experiment questionnaire that was intended to determine whether they recognized a phonological re lation between the pictures and the target words, and if so, whether they intentionally used this association to determine the target word. Participants were then debriefed. Results The responses to the post-experiment ques tionnaire revealed th at no participants recognized the intended relations hip between the questions and pr ime pictures and thus did not consciously use the relationship to influence their responses to the questions and pictures, a finding consistent with previous TOT research (e.g., Abrams et al., in press) All analyses were 26

PAGE 27

conducted by participants only (to get a mean for each participant rather than each item) because the stimulus items were not selected randoml y, a condition that has be en suggested to be inappropriate for item analyses (e.g., Cohen, 1976; Raaijmakers, Schrijnemakers, & Gremmen, 1999; Vitevitch & Sommers, 2003). Response Types for Questions, Reported Minutes for TOT Resolution, and Picture Names Of the initial responses to the questions, 35.7% were co rrect "Known" responses, 15.6% were incorrect "Known" respons es, 10.3% were correct "TOT" responses, 1.9% were incorrect "TOT" responses, 25.3% were "U nknown" responses, and 11.1% were "None of the Above" responses. A correct "Known" response was when the answer given to the general knowledge question matched the target. A correct "TOT" response was one in which the participant correctly retrieved the target on the stem-completion test or chose the correct answer on the recognition test. After a "TOT" res ponse, participants estimated th at they could resolve the TOT within 10 minutes on 94.4% of the responses ( M = 4.95 minutes, SD = 4.57, range = 1-40). Of the responses to the prime pictures, 76.7% produced the intended dominant name, 14.3% produced the intended secondary name, and 9.1% produced names other than the intended dominant and secondary names. Stem-Completion Test Because the stem-completion test determined if participants could retrieve the word when provided with letter cues, it examined the activati on level of the target word for participants who were and were not having a TOT. To investigat e the efficacy of the stem-completion test, for initial correct "TOT" responses, "Unknown" responses, and "None of the Above" responses, a repeated-measures ANOVA was conducted on mean pe rcentage of targets correctly retrieved on the stem-completion test. The main eff ect of response type was significant, F (2, 88) = 16.90, 27

PAGE 28

MSE = .04, p < .001, with more correct target retr ievals following "TOT" responses ( M = 66.13%) than either "Unknown" responses (M = 44.68%) or "None of the Above" responses ( M = 44.44%), which did not differ (p > .95). Because of the nonsi gnificant difference between "Unknown" and "None of the Above responses, we combined these responses into one category for subsequent analyses. Mediated Priming To test the hypothesis that mediated priming would influe nce TOT incidence, pairedsamples t-tests with prime condition (Prime Pictur e, Control Picture) we re conducted separately on the proportion of correct "Know n" responses (i.e., proportion of targets correctly retrieved), correct "TOT" responses (i.e., proportion of corr ect TOT states), and "U nknown"/"None of the Above" responses. Only non-secondary name producti ons (i.e., dominant na mes as well as other name productions that did not overlap phonologically with the targ et) of the prime picture were included in the analyses, as those reflected medi ated priming; the secondary name overlapped in first-syllable phonology and therefore would be a direct phonological prim e for the target if produced. Means and standard deviations in percen ts are presented in Table 2-1. T-tests revealed significant effects of prime condition. Th ere were more "Known" responses, t (59) = 4.21, SE = .03, p < .001, fewer "TOT" responses, t (59) = -2.16, SE = .01, p < .04, and marginally fewer "Unknown"/"None of the Above" responses, t (59) = -1.68, SE = .02, p < .10, following prime pictures than control pictures. Paired-samples t-tests were th en conducted on the proportion of each response type following only dominant name productions to clarify whether including other phonologically-unrelated responses in analysis were contributing to the above results. These ttests revealed similar results to those given above, with more "Known" responses, t (59) = 4.37, SE = .02, p < .001, marginally fewer "TOT" responses, t (59) = -1.79, SE = .01, p < .08, and 28

PAGE 29

fewer "Unknown"/"None of the Above" responses, t (59) = -2.07, SE = .02, p < .05, following prime pictures. Direct Priming Because of its overlapping phonology with the ta rget, production of the prime pictures secondary name was a direct prime of the targ et. Therefore, direct priming on TOT incidence was examined via paired-samples t-tests with pr ime condition (Prime Picture, Control Picture) that were conducted separately on the proportion of correct "Known" re sponses (i.e., proportion of targets correctly retr ieved), correct "TOT" responses (i.e ., proportion of correct TOT states), and "Unknown"/"None of the A bove" responses. In this analysis, only secondary name productions of the prime picture we re included in the analyses, as their first-syllable matched that of the target. Mean s and standard deviations in percents are presented in Table 2-2. Unlike the previous results on mediated priming, t-test s showed equivalent pr oportions of correct "Known" responses, t < 1, correct "TOT" responses, t < 1, and "Unknown"/"None of the Above" responses, t < 1, following production of the prime pictur es secondary name versus the control picture. Part of Speech An analysis was conducted to determine if th e observed mediated priming effects were dependent on the prime and target's part of speech, i.e., whether more priming occurred when primes and targets differed in part of speech than when they shared part of speech. Repeatedmeasures ANOVAs were conducted separately on percentage of corr ect "Known" responses, correct "TOT" responses, and "Unknown"/"None of the Above" responses, using Prime Part of Speech as a factor (phonologically-mediated prime and target were the same part of speech, phonologically-mediated prime and target were diffe rent parts of speech, and unrelated control 29

PAGE 30

picture that was the same part of speech as the prime). Means and standard deviations in percents from the ANOVAs are presented in Table 2-3. For "Known" responses, there was a signifi cant effect of Prime Part of Speech, F (2, 118) = 8.34, MSE = .02, p < .001, with more "Known" responses following same part-of-speech primes and different part-of-speech primes than control pictures, ps < .03. There were also marginally more "Known" responses following diffe rent part-of-speech primes than same partof-speech primes, p < .10. For "TOT" responses, the Prim e Part of Speech effect was not significant, F (2, 118) = 1.85, MSE = .01, p > .16. Similarly, for "Unknown"/"None of the Above" responses, there was no effect of Prime Part of Speech, F (2, 118) = 1.12, MSE = .01, p > .33. Discussion As predicted, mediated priming influen ced TOT incidence, demonstrated by more Known responses and fewer TOT responses following prime pictures relative to control pictures. Production of the prime picture's dominant name activated the first-syllable phonology of the upcoming target via mediated priming, allowing part icipants to retrieve the target more often than when the first-syllable phonology was not activated. Even though the act ivation strength of the phonology via mediated priming is thought to be weaker than that of direct priming (Dell & OSeaghdha, 1991, 1992), the findings suggest that th e activation is strong enough to strengthen the targets phonological connec tions, inducing it to be more accessible for production. Despite finding mediated priming, we found no evidence for direct priming on incidence of TOT states, as production of the secondary name that overlapped phonologically with the target did not reduce TOT states or increase Know n responses. These results are contradictory to previous findings that direct priming decrea ses TOT incidence (James & Burke, 2000), but the 30

PAGE 31

current findings may not be a true assessment of direct primings influence on TOT incidence because of the low percentage of secondary name productions (14.3%), which created a large amount of variability in participants' means and may have redu ced the statistical power to detect an effect of direct priming. The mediated priming effects on TOT incide nce occurred independent of the overlap between the prime and target's parts of speech. Previous research has found that a direct primes part of speech (i.e., whether it shares the same or has a different part of speech than the target) affects TOT resolution differently (Abrams & Ro driguez, 2005; Abrams et al., in press); different part-of-speech primes increase TOT resolution, whereas same part-of-speech primes have no effect. The current study also found effects due to the mediated prime's part of speech on TOT incidence, but both sameand different part-o f-speech primes facilitated target retrieval by increasing "Known" responses, relative to control pictures, and different part-of-speech primes helped marginally more than same part-of-sp eech primes. While the different part-of-speech findings were predicted by the TDH, the facilitatory effect of same part-of-speech primes was not. The present study involved TOT incidence, wher eas the previous studies on part of speech have only investigated their e ffects on TOT resolution, so perhaps TOT resolution is more sensitive to grammatical cl ass than TOT incidence. In sum, these findings demonstrated another process by which TOT states can be prevented from occurring in the first place. In addition to direct exposure to a phonologicallysimilar prime (James & Burke, 2000), TOT incidence may also be lowered via mediated priming, as the activation sent to related associates is strong enough to activate the node above threshold for production. Experiment 2 investigat ed mediated priming effects on TOT resolution. While TOT incidence and TOT resolution both involve strength ening the phonological 31

PAGE 32

32 connections to the selected word for producti on, decreasing incidence re quires strengthening prior to lemma selection, whereas increasing resolution requires stre ngthening after lemma selection. Determining if mediated priming can affect TOT incidence a nd resolution similar, similar to findings with direct priming (James & Burke, 2000), will provide further understanding of the processes involved in strengthening th e phonological connections to both prevent and resolve TOT states.

PAGE 33

Table 2-1. Initial responses (i n %) following producti on of the prime pictures non-secondary name and control pictures' name in Experiment 1 ________________________ _____________________ _________________________ Prime Condition N Prime Picture Control Picture Initial Response Mean SD Mean SD ______________________________________________________________________________ Known 60 42.21 19.23 31.66 15.35 TOT 60 8.75 7.55 11.73 9.03 Unknown/None of the Above 60 34.65 17.25 37.63 16.23 ______________________________________________________________________________ 33

PAGE 34

Table 2-2. Initial responses (in %) following production of the pr ime pictures secondary name and control pictures' name in Experiment 1 ________________________ _____________________ _________________________ Prime Condition N Prime Picture Control Picture Initial Response Mean SD Mean SD ______________________________________________________________________________ Known 60 33.07 25.66 31.33 15.41 TOT 60 11.07 13.83 11.80 8.96 Unknown/None of the Above 60 38.80 25.13 37.65 16.15 34

PAGE 35

Table 2-3. Initial responses (in %) following prime pictures that shar ed the part of speech (POS) with the target, prime pictures that differed in pa rt of speech from the target, and control pictures in Experiment 1 ____________________________________________________________________________________ Prime Condition N Same POS Prime Different POS Prime Control Picture Initial Response Mean SD Mean SD Mean SD _____________________________________________________________________________________________ Known 60 37.99 21.49 41.75 20.20 31.33 15.41 TOT 60 10.00 9.98 8.96 8.59 11.80 8.96 Unknown/None of the Above 60 35.27 17.83 35.07 18.11 37.65 16.15 ______________________________________________________________________________ 35

PAGE 36

36 Biopsy (if Known) What noun means the removal of a sample of tissue, which is then examined under a microscope to check for cancer cells? Sheep (Lamb) Motorcycle (Bike) Anchor Respond: Known, TOT, Unknown, or None of the Above View Question Name Pictures Figure 2-1. Order of tasks for Experiment 1. Participants named three pictures (two filler pictures and either a prime or cont rol picture) and then answered a general knowledge question. The question was associ ated with a target word that was phonologically related to the secondary name of the prime picture. The target names of the pictures are below the correspondi ng picture. The secondary name of the prime picture is in parentheses.

PAGE 37

CHAPTER 3 EXPERIMENT 2 Experiment 2 was designed to examine wh ether mediated priming influences TOT resolution, i.e., retrieval of targ ets following TOT states. Because mediated priming effectively influenced TOT incidence in Experiment 1, i.e ., because the activation from the dominant name (e.g., motorcycle ) to the secondary name (e.g., bike ) was strong enough to activate the phonology for the target word to decrease TOT incidence, we expected the same st rength of activation to facilitate TOT resolution. In a ddition to mediated priming, we al so investigated differences in resolution following direct priming (production of the secondary name). Because direct priming has been shown to successfully facilitate TOT resolution (e.g., Abrams et al., 2003; Abrams et al., in press), we predicted grea ter resolution following direct pr iming relative to no priming, as the target receives more activation directly from a prime rather than indirectly (Dell & OSeaghdha, 1991, 1992). A secondary examination in this experiment wa s the ability of a TOT word to feed back activation to influence lemma sel ection; that is, can having a TO T induce participants to produce the secondary name of the pict ure, the phonologically-related name to the TOT word, more often than they would otherwise? We expected more secondary name productions when participants were in a TOT state (i.e., the target activat es the secondary names phonology via bottom-up processing) relative to when they did not know th e answer (i.e., the target is not activated, so bottom-up processing cannot occur to activate secondary names phonology). Method Participants Participants included 60 young adults ( 41 females, 19 males), aged 18-25 ( M = 18.92, SD = 1.13), recruited from introductory psychology c ourses and given partial course credit for 37

PAGE 38

participation. All considered English to be their native language and did not participate in Experiment 1. Materials The same materials in Experiment 1 were used in Experiment 2. Design and Procedure The experimental design and procedure for Experiment 1 was used for Experiment 2, except for the order of the tasks. Th at is, participants named a picture after their response to the questions (see Figure 3-1 for example). Thus, participants viewed a question associated with a low-frequency word and responded "Known", "Unknown", "TOT", or "None of the Above". After a "Known" response, participants typed their response, named a control and two filler pictures, and proceeded to the next questi on. After a "TOT", "Unknown", or "None of the Above" response, participants named two filler pictur es and either a prime or control picture, and then attempted to answer the same question before proceeding to the next question. Results Similar to Experiment 1, no participants i ndicated on the post-experiment questionnaire that they recognized the intended relationship between the questions a nd prime pictures or consciously used the relationship to influence their responses to the questions and pictures. Response Types for Questions, Reported Minutes for TOT Resolution, and Picture Names Of the initial responses to the questions, 36.8% were co rrect "Known" responses, 15.1% were incorrect "Known" respons es, 11.1% were correct "TOT" responses, 1.9% were incorrect "TOT" responses, 24.9% were "U nknown" responses, and 10.3% were "None of the Above" responses. After a "TOT" response, participants es timated that they could resolve the TOT within 10 minutes on 91.8% of the responses ( M = 5.43 minutes, SD = 6.57, range = 1-43). Of the 38

PAGE 39

responses to the prime pictures, 70.7% were the intended dominant name, 20.9% were the intended secondary name, and 8.4% were names othe r than the intended dominant and secondary names. Stem-Completion Test Similar to Experiment 1, the stem-completion test examined the activation level of the target word for participants who were and we re not having a TOT. Fo r initial correct "TOT" responses, "Unknown" responses, and "None of the Above" responses, a repeated-measures ANOVA was conducted on mean percentage of correctly retrieved targets on the stemcompletion test. Fourteen participants were excl uded from this analysis because they did not have to take the stem-completion test at least once following each response type. The main effect of response type was significant, F (2, 90) = 15.84, MSE = .02, p < .001, with more correct target retrievals following "TOT" responses (66.21 %) than "Unknown" (44.41%) and "None of the Above" (46.18%) responses, which did not differ from each other ( p > .66). Because of the nonsignificant difference between "Unknown" and "None of the Above" responses, these responses were combined in subsequent analyses. Mediated Priming A 2 (Response Type: correct "TOT", "Unknow n"/"None of the Above") x 2 (Prime Condition: Prime Picture, Control Picture) repeated-measures ANO VA was conducted on the proportion of targets correctly retrieved following the pictur e naming task. A correct "TOT" response was one in which the participant res ponded "TOT" when they first saw the question, and either 1) responded "Known" and provided th e correct answer when viewing the question a second time; 2) provided the correct answer on the stem-completion test; 3) correctly identified the answer on the recognition test. This anal ysis included only non-secondary name productions 39

PAGE 40

(dominant and other phonologically-unrelated responses), as those productions measure mediated priming, whereas secondary name productions assessed di rect priming. Four participants were excluded from this analysis because they did not provide the non-secondary name of the prime picture at least in once in all of the c onditions. The means and standard deviations from this analysis converted into percents are shown in Table 3-1. The main effect of prim e condition was significant, F (1, 55) = 22.19, MSE = .12, p < .001, showing more correct target re trievals following prime pictur es than control pictures. The main effect of response t ype was also significant, F (1, 55) = 50.03, MSE = .06, p < .001, with greater resolution after correct "TOT" respons es than after "Unknown"/"None of the Above" responses. These main effects were modera ted by a significant Response Type x Prime Condition interaction, F (1, 55) = 5.39, MSE = .04, p < .03, shown in Figure 3-2. Follow-up tests on the Response Type x Prime Cond ition interaction revealed that significant priming occurred after initial correct "TOT" responses, p < .001, with more correct ta rget retrievals following prime pictures than control pictures. There was also significant priming after initial "Unknown"/"None of the Above" responses, p < .001, but to a lesser ex tent. There was greater target retrieval following correct "TOT" re sponses than "Unknown" /"None of the Above" responses for both prime conditions, with a larger difference for the prime pictures, ps < .001. To assess whether including other phonologi cally-unrelated responses in the above analyses contributed to the fi ndings, a 2 (Response Type: correct "TOT", "Unknown"/"None of the Above") x 2 (Prime Condition: Prime Picture, Control Picture) repeated-measures ANOVA was conducted on target retrieval following dominant name producti ons only. Seven participants were excluded from the analysis because they di d not have at least one intended dominant name 40

PAGE 41

production in all of the conditions. The analysis revealed identical results to the previous analysis. Direct Priming A 2 (Response Type: correct "TOT", "Unknow n"/"None of the Above") x 2 (Prime Condition: Secondary Name Production of Prime Pi cture, Control Picture) repeated-measures ANOVA was conducted on the propor tion of targets corr ectly retrieved fo llowing the picture naming task. Thirty-three particip ants were excluded from this analysis because they did not provide the secondary name of the prime pictur e at least once following both response types. The means and standard deviations from this analysis converted into percents are shown in Table 3-2. The main effect of prim e condition was significant, F (1, 26) = 24.58, MSE = .23, p < .001, showing more correct target retrievals following secondary name productions of prime pictures than control pictures. The main e ffect of response type was also significant, F (1, 26) = 16.09, MSE = .07, p < .001, with greater resolution af ter correct "TOT" responses than after "Unknown"/"None of the Above" re sponses. These main effects we re moderated by a marginally significant Response Type x Prime Condition interaction, F (1, 26) = 2.98, MSE = .04, p < .10. Follow-up tests on the Response Type x Prime C ondition interaction revealed that significant priming occurred after initial correct "TOT" responses, p < .0 01, with more correct target retrievals following secondary name productions of prime pictures than control pictures. There was also significant priming after initial "U nknown"/"None of the A bove" responses, p < .001, but to a lesser extent. There wa s greater target retrieval followi ng correct "TOT" responses than "Unknown"/"None of the Above" re sponses for both prime conditions, with a larger difference for the prime pictures, ps < .001. 41

PAGE 42

To compare whether there was greater primi ng of target retrieval following secondary versus non-secondary name production, a 2 (Res ponse Type: correct "TOT", "Unknown"/"None of the Above") x 3 (Prime Condition: Seconda ry Name Production of Prime Picture, NonSecondary Name Production of Prim e Picture, Control Picture) was conducted on the proportion of targets correctly retrieved. Thirty-three particip ants were excluded from this analysis because they did not have at least one correct "TOT" response and on e "Unknown"/"None of the Above" response in each of the prime conditions. The ma in effect of response type was significant, F (1, 26) = 26.79, MSE = .07, p < .001, with more correct target retrievals following correct "TOT" responses (44.6%) than "Unknown"/"None of th e Above" (23.8%) responses. Additionally, the main effect of prime condition was significant, F (2, 52) = 20.57, MSE = .14, p < .001, where both secondary and non-secondary name production led to greater target retrieval than the control picture (8.2%), p s < .001, indicating significant prim ing for both types of names. However, more correct target retrievals o ccurred following secondary name production (52.4%) than non-secondary name production (42.1%), p < .005. The Response Type x Prime Condition interaction was not significant, F (2, 52) = 1.62, MSE = .04, p > .20. Part of Speech Potential part of speech effects on mediated priming were examined via a 2 (Response Type: Correct "TOT", "Unknown"/"None of th e Above") x 3 (Prime Part of Speech: phonologically-mediated prime and target were the same part of speech, phonologicallymediated prime and target were different parts of speech, and unrelated co ntrol picture that was the same part of speech as the prime) repeated-measures ANOVA on target retrieval. Twenty participants were excluded from the analysis because they did not have at least one correct "TOT" response and "Unknown"/"None of the Above response in each of the prime part-of42

PAGE 43

speech categories. The means and standard deviations from this analysis converted into percents are shown in Table 3-3. The main effect of response type was significant, F (1, 39) = 53.58, MSE = .07, p < .001, showing more correct target re trievals following correct "TOT" than "Unknown"/"None of the Above" responses. The main effect of prime part of speech was also significant, F (2, 78) = 16.01, MSE = .13, p < .001, with greater resolution after same and different part-of-speech prim es than control pictures, p < .001, but no difference in target retrieval between same and diffe rent part-of-speech primes, p > .15. The Response Type x Prime Part of Speech interaction, F (2, 78) = 1.60, MSE = .05, p > .20, was not significant. Feedback To examine if the TOT word can feed back to influence picture name production, a paired-samples t-test was conducted on percen tage of secondary name productions following correct "TOT" responses and "Unknown"/"None of the Above" responses. There was no difference between secondary name produc tions after correct "TOT" responses ( M = 19.8%) and "Unknown"/"None of the Above" responses ( M = 20.2%), t < 1. The feedback effect was also examined by comparing percentage of secondary name productions in Experiment 2 with those in Expe riment 1 (where the picture was named first before viewing the questions, so there was no opportunity for feedback from the unretrieved target). Independent-samples t-tests were conducted on perc entage of secondary name productions between the two experiments, se parately for correct "TOT" responses and "Unknown"/"None of the Above" responses. Foll owing "TOT" responses, there were more secondary name productions in Experiment 2 (19.8%) than in Experiment 1 (14.5%), t (117) = 1.74, SE = .03, p < .001. A similar pattern emerged for secondary name productions following 43

PAGE 44

"Unknown/None of the Above" responses in Expe riment 2 (20.2%) than Experiment 1(14.5%), t (118) = -2.49, SE = .02, p < .001. Discussion Mediated priming increased TOT resoluti on by strengthening the needed phonological connections for target retrieval, evidenced by gr eater target retrieval following production of the prime pictures' dominant name relative to produc tion of control pictures. Even though the firstsyllable phonology of the TOT word was only in directly activated (i.e., production of the dominant name activated its semantic associat e, the secondary name, which spread phonological activation to the TOT word), the activation strength was st rong enough to strengthen the phonological connections for TOT resolution. Similar to TOT incidence, this mediated priming effect was independent of part of speech, as both same and different part-of-speech primes facilitated TOT resolution. The hypothesis that direct priming would influence TOT resolution was also supported, as production of prime pict ures' secondary name increased TOT resolution relative to control pictures. In addition to replic ating previous findings on the facilitation of TOT resolution via direct pho nological priming (e.g., Abrams et al., 2003; Abrams et al., in press; James & Burke, 2000), the current study extends thes e findings by showing that direct priming is more effective for resolving TOTs than mediated priming. In addition to TOT states, direct and medi ated priming also unexpectedly facilitated target retrieval after "Unknown"/"None of the Above" response s, a contradictory finding from previous TOT research that has shown virtually no target retrie val following "Unknown" responses (Abrams & Rodriguez, 2005; Abra ms et al., 2006; James & Burke, 2000). One methodological difference between previous studie s and the current study is that we included participants' estimates of their TOT resolution time. Participants in both Experiment 1 and 2 44

PAGE 45

estimated that they could resolv e the majority of their TOT stat es (over 90% of them) within 10 minutes. One possibility is that asking participan ts to report the time to resolve their TOT may have caused them to be more conservative about cl assifying a state a TOT state. Therefore, they classified only strong TOT states, ones in whic h they could resolve w ithin 10 minutes, as TOT states; weaker TOT states with a less sense of im minent retrieval may have instead been reported as "Unknown"/"None of the Above" responses. Benefiting from priming, the TOT states included in the Unknown/None of the Above cate gory would cause the cate gory to show priming effects. The results from mediated primin g's influence on TOT in cidence supports this explanation, as fewer "Unknown"/"None of the Above" responses occurred following prime than control pictures. If participants were in a true "unknown" state, there should be no change in these responses following priming. Furthermore, the results from the stem-completion test support this misclassification explan ation, as targets were retrieved at a fairly high rate following "Unknown" (44.14%) and "None of the Above (46.18%) responses, although not as high following "TOT" responses (66.21%); if participants had responded "Unknown" only when the target was truly unknown, the retrieva l rate on the stem-completion te st would be expected to be closer to 0%. Additionally, despite its priming effect, the mediated and direct priming effects for "Unknown"/"None of the Above" responses were sm aller than those for "TOT" responses, which supports the misclassification theory. Because th e activation strength of both the TOT state and the prime determines if the TOT is activated above threshold for production, weak TOT states will need more activation strength from primes to get resolved than strong TOT states. Therefore, because the TOT states in "Unknow n"/"None of the Above" and "TOT" responses received the same primes, i.e., received the same activation, then the pr iming effect difference 45

PAGE 46

between "TOT" and "Unknown"/"None of the Above responses must be attributable to the activation strength of the TOT states. The prediction that secondary name produc tions would be more prevalent when participants were in a TOT state than when in an "unknown" state, as the activated TOT word would feed back to activate the phonology of the secondary name enough for production, was not supported. Following correct "TOT" responses, participants were no more likely to produce the secondary name than after "Unknown"/"None of th e Above" responses, suggesting that the bottom-up processing from TOT states was not st rong enough to influence lemma selection. An alternative explanation is that the lack of secondary name productions follow TOT and "Unknown"/ "None of the Above" responses could be due to the picture naming task; that is, the secondary name could always be activated for retrieval, an activation that may be greater than what is experienced in the TOT state. Theref ore, regardless of whether in a TOT state, participants may produce the secondary name mere ly because it is always activated with the dominant name. However, the comparison of s econdary name productions in Experiment 1 and Experiment 2 does demonstrate the occurrence of a feedback effect. Becau se participants in Experiment 1 named the pictures before viewi ng the question, the fee dback effect was not possible; thus, Experiment 1 served as a contro l group to which secondary name productions in Experiment 2 were compared. The feedback e ffect was possible in Experiment 2 because participants named the pictur e after viewing the question. For both TOT and "Unknown"/"None of the Above responses, partic ipants were more likely to produce the secondary name in Experiment 2 than Experiment 1. In the cases in which participants produced the secondary name in Experiment 2, th e activation to the secondary name was strong enough to exceed that of the dominant name, the most frequently produced name, to 46

PAGE 47

become the most-primed-wins node and get produ ced. Because this effect also occurred when participants said they were in an "unknown" state, we cannot decisively conclude that the activated TOT lemma is responsible for the effect. However, if participants misclassified th eir weaker TOTs as "unknown" states as suggested earlier, then the part ial activation of TOT states th at would occur in both TOT and Unknown/None of Above categories may be strong enough to transmit enough phonological activation to related lemmas fo r production. If this partial activ ation of TOTs explanation is correct, then the feedback eff ect should have been smaller for "Unknown"/"None of the Above" responses than "TOT" responses, si milar to the smaller effects of direct and mediated priming on "Unknown"/"None of the Above" re lative to "TOT" responses. The finding that the feedback effect was similar for both response types sugge sts that the phonological activation strength of weak TOTs was sufficient to influence lemma se lection similar to that of strong TOTs. Thus, misclassified TOTs, i.e., weak TOTs, may infl uence TOT resolution and feedback differently: More phonological activation may be required for a word to emerge from a TOT state (TOT resolution) than for a word to exceed the activ ation of the more freque ntly-produced dominant name (feedback). Thus, the activat ion strength of TOTs seems mo re important in TOT resolution than feedback. In TOT resolution, the word b ecame a TOT state becaus e of weak phonological connections (Burke et al., 1991; MacKay, 1987). For the word to be retrieved from its TOT state, the phonological activation had to exceed threshold. Because of their greate r activation strength, strong TOTs needed less activation from the primes than weak TOTs; thus, TOT resolution occurred more for strong than weak TOTs. In the feedback instances, the common, familiar secondary name was different from the TOT wo rd because their phonological connections were strong enough to not become a TOT state. Ther efore, the secondary na me was activated enough 47

PAGE 48

when participants viewed the picture to onl y need a small amount of additional phonological activation from the TOT lemma to become the most-primed-wins node over the dominant name, meaning that the phonological activ ation from the weak TOT may be sufficient for the name to be produced (strong TOTs just provide extra activa tion that is not needed ). In turn, TOTs may influence the lexical system sometimes better than they can be influenced. In sum, similar to its influence on TOT in cidence, mediated priming facilitated TOT resolution by indirectly activa ting the first-syllable phonology of the TOT word above threshold for production. These findings demonstrate that even though the activation st rength sent to target words via mediated priming is weaker than that of direct priming (Dell & OSeaghdha 1991, 1992), it is sufficient to influence TOT states. T hus, comparable to direct priming (e.g., Abrams et al., in press; James & Burke, 2000), mediat ed priming is an effective priming method by which spontaneous TOTs may be resolved. Collec tively, these studies dem onstrate that mediated priming can benefit TOT states by decreas ing incidence and increasing resolution. 48

PAGE 49

Table 3-1. Target retrieval (i n %) following production of th e prime pictures non-secondary name and control pictures' name in Experiment 2 ________________________ _____________________ _________________________ Prime Condition N Prime Picture Control Picture Initial Response Mean SD Mean SD ______________________________________________________________________________ TOT 56 46.63 40.39 18.64 28.76 Unknown/None of the Above 56 18.36 27.87 2.37 5.10 ______________________________________________________________________________ 49

PAGE 50

Table 3-2. Target retrieval (in %) following production of the prime pictures secondary name and control pictures' name in Experiment 2 ________________________ _____________________ _________________________ Prime Condition N Prime Picture Control Picture Initial Response Mean SD Mean SD ______________________________________________________________________________ TOT 27 65.93 45.60 14.82 24.02 Unknown/None of the Above 27 38.91 44.07 1.57 3.72 ______________________________________________________________________________ 50

PAGE 51

Table 3-3. Target retrieval (in %) following prime pi ctures that shared the part of speech (POS) with the target, prime pictures that differed in pa rt of speech from the target, and control pictures in Experiment 2 ________________________ _____________________ _________________________ Prime Condition N Same POS Prime Different POS Prime Control Picture Initial Response Mean SD Mean SD Mean SD ______________________________________________________________________________ TOT 40 55.38 40.71 50.05 45.43 19.85 29.24 Unknown/None of the Above 40 26.50 36.12 21.21 34.43 1.88 4.11 ______________________________________________________________________________ 51

PAGE 52

52 What noun refers to a system or condition of society in which the means of production are owned and controlled by the state? Socialism (if Known) What noun refers to a system or condition of society in which the means of production are owned and controlled by the state? Necklace (Pearls) Couch (Sofa) Tree Respond: Known, TOT, Unknown, or None of the Above View Question (if Response was TOT, Unknown, or None of the Above) View Question Name Pictures Figure 3-1. Order of tasks for Experiment 2. Pa rticipants viewed a question, provided their response, and then named three pictures (two filler pictures and either a prime or control picture). The question was associ ated with a target word that was phonologically related to the secondary name of the prime picture. The target names of the pictures are below the co rresponding picture. The secondary name of the prime picture is in parentheses.

PAGE 53

53 0 10 20 30 40 50 60 Response Type% Correct Target RetrievalCorrect TOT Unknown/None of the Above Control Picture Prime Picture Figure 3-2 Percent correct target retrieval following pr ime and control pictures for "TOT" and "Unknown"/"None of the Above" responses in Experiment 2.

PAGE 54

CHAPTER 4 GENERAL DISCUSSION Analogous to Peterson and Savoy's (1998) demonstration of effective mediated priming in word recognition the current study extended this phenomenon to word retrieval in relation to TOT states. These results coincide with prev ious TOT evidence that support interactive activation models of language production (Dell, 1986; MacKay, 1987; Stemberger, 1985), thereby supporting Dell et al.s (1997) hypothesis that interactiv e activation is beneficial for fluent discourse. Because the phonological activati on can occur before the selection of a lemma (i.e., the most activated lemma to be selected as the most-primed-wins node), discourse is more fluent as the nodes can be prepared for production via priming. In the case of TOT incidence, the lemma needed for production can receive phonological activation via mediated priming before lemma selection, thereby decreasing the incidence of TOT states. If the lemma does not receive sufficient activation prior to selection, it can nevertheless receive additional priming after selection, even if the individual is exposed to the phonology only i ndirectly. Therefore, mediated priming in TOT states furthers the notion that bottom-up processing is an important part of the lexical system, a process that is not pred icted by discrete theo ries (Levelt, 1989). In addition to interactive activation models, th e findings that mediated priming influences TOT states supports the recency hypothesis of the TDH (Burke, et al., 2000; Burke et al., 1991; MacKay & Burke, 1990). Because the target 's phonological connections were recently strengthened via the i ndirect activation of the phonologically-related secondary name, mediated priming decreased TOT incidence and increased TOT resolution, similar to the process thought to underlie direct priming from phonologica lly-related words (James & Burke, 2000). In conclusion, mediated priming has parallel effects on TOT inciden ce and resolution, as the same processes to strengthen phonologica l connections help to reduce as well as resolve TOT states. 54

PAGE 55

Replicating previous TOT research (Abrams & Rodriguez, 2005; Abrams et al., in press; Abrams et al., 2003), the current study found that direct phonological primi ng of the TOT words first syllable facilitated TOT resolution. A new contribution to the understanding of TOT states is that direct priming is more effective in re solution than mediated priming, as was predicted by Dell and OSeaghdha (1991, 1992). Naming the pi cture's secondary name provided more activation to the phonological connections, thereby providing a greater chance of TOT resolution, because the phonology was a direct recipi ent of the activation. In mediated priming, the phonology is an indirect recipient that receiv ed a proportion of the original activation; thus, the probability of the TOT state receiving enough activation for production decreased. Although not explicitly manipulated, the curr ent study found part-ofspeech to affect mediated priming of TOT incidenc e and resolution similarly, wher e both different and same partof-speech primes decreased TOT incidence (measured by an increase in "Known" responses) and increased TOT resolution. However, while di fferent part-of-speech primes decreased TOT incidence more than same part-of-speech prim es, there was no difference in TOT resolution following same and different part -of-speech primes. These findings contrast with Abrams and Rodriguez (2005) and Abrams et al. (in press), who found that primes of a different part-ofspeech from the TOT word facil itated resolution, whereas primes of the same part-of-speech resolution did not increase target retrieval. The similar, facilitatory effects of same and different part-of-speech primes on TOT states in the pres ent study may be attributable to the frequency of the secondary names of the primes (i.e., freque ncy of picture name productions given as the second response on the pilot picture naming task; M = 38.80), which was lower than those in Abrams et al. (2006; M = 63.01). Frequency of same part-of-speech primes has been shown to negatively correlate with TOT resolution (Abrams & Rodriguez, 2005). Therefore, the primes 55

PAGE 56

may have been too low in frequency to serve as potential competitors with the target for retrieval, which is necessary to delay retrieval as explained by the "most primed wins" principle of the Node Structure Theory (MacKay, 1987). Th e activation of a node in the same syntactic class as the target word should not help retrieval because the activated prime would be the mostprimed-wins node for that class. However, not being activated as much as higher-frequency primes, low-frequency primes are activated enough to provide phonological activation to the target word but not compete to be the most-primed-wins node. Furthe rmore, as Dell and OSeaghdha (1991, 1992) explained, mediated primes are activated less and, thus, send less activation to the target than dire ct primes because they receive only a proportion of the original activation from their activated associate. Becau se they do not have enough activation to be activated above threshold for production, mediated primes may not be as strong of a competitor in their grammatical class to be the most-primed-wins node as direct primes. Therefore, a target word in the same grammatical class can receive phonological activation from the mediated prime but not have to compete with the partially-activated prime. The feedback effect was demonstrated thr ough TOT states, as the partially-activated words influenced lemma selection. Additionally, if participants misclassified their weak TOT states in the Unknown/None of th e Above category, the results reve al that weak TOT states may have the same influence on the lexical systems as strong TOT states, which the further support the interactive activation theories, as botto m-up processing was required for the activated phonology to influence lemma selection. Additiona lly, these findings further our understanding of TOT states and their role in the lexical syst em. However, further investigations are necessary to clarify the relationship between the TOT state and the feedback effect. Instead of measuring the feedback effect by secondary name productions of near-synonymous pictures, future research 56

PAGE 57

could investigate picture naming times following TOT and unknown states. If TOT states can uniquely feed back to prime lemmas, then partic ipants should name the pictures faster following TOT states than unknown states. To specifically to determine if weak TOT states can influence lemma selection similarly to strong TOT states, future research could similarly investigate picture naming times but distinguish between weak and strong TOTs based on feeling-ofknowing judgments. If the TOT state does influence lemma sel ection, the finding will not only demonstrate the activation strength of TOT states but also their role in the lexical selection (i.e., similar to the way in which they receive phonol ogical priming, they can provide phonological priming to related lemmas). Addi tionally, the feedback effect could be used understand the duration of TOT states: Can the f eedback effect occur for TOT stat es that have been unresolved for 20 minutes similar to those that have b een unresolved for 5 minutes? Theoretically, the sooner the TOT onset to lemma selection of a nother word, the most activation the TOT word and, thus, greater influence it wi ll have on the lexical system. Whereas this experiment investigated the TDH's recency hypothesis, future research might investigate the TDH's agi ng hypothesis by examining age-rela ted influences of mediated priming on TOT states. Because aging weak ens phonological connections (e.g., MacKay & Burke, 1990), node priming diminishes with age. In mediated priming, the TOT state being activated above threshold for production relies on a function of the activation of distant associates and the strength of the phonological connections. Thus, if one of these variables is weak, as the latter one is in aging, then medi ated priming cannot activate the TOT word above threshold, which may be seen in older adults, part icularly old-old adults. These results may have several informative outcomes. First, they would coincide with previous evidence demonstrating age-related changes in direct phonological priming of TOT resolution (Abrams et al., in press; 57

PAGE 58

White & Abrams, 2002) and offer support for the TDH's aging hypothesis. Second, the results would help explain why older adults experien ce more TOT states than younger adults: Younger adults benefit from both direct and mediated pr iming to decrease TOT incidence, whereas older adults may only benefit from direct priming. Third, they would demonstrate that mediated priming may not be an effective method for inve stigating cognitive processes in older adults. The previously discussed possibi lity of the misclassification of weak TOT states is an evident limitation to the current study. Although the results demonstrate that mediated priming is an effective priming strategy, we cannot decisively conclude that the feedback effect is due to TOT states. Additionally, while we attempted to include clear near-synonymous pictures, i.e., ones that have a clear dominant and secondary name that mean the same thing, we had to resort to using some that were more of se mantic associates than near-synonyms (e.g., world and globe ) to increase the stimuli size. However, nea r-synonyms and semantic associates may differ lexically, which could affect their role in mediated priming. For example, mailman and postman are clear near-synonyms, they can be appropriately used interchangeably and have the exact same definitions. Therefore, when ei ther the definition or picture of mailman / postman is presented, theoretically the node representing th e definition should send equal activation to both mailman and postman. Receiving equal activation as it near -synonym, the node that had the most activation prior to the definition or picture being presented, e.g., mailman, is the one that is selected for production. However, being th e secondary name, the other node, e.g., postman, can still influence the lexical system as it is partially activated. Conversely, the node representing the definition or picture of a globe should only send activation to globe as it is the only term that accurately fits the definition or describes the pi cture. Because they are semantic associates, globe can then send activation to world ; however, because world did not receive its activation from the 58

PAGE 59

original source, i.e., th e picture of the globe, globe will have more activation strength than world. Therefore, postman the secondary name in the near-synony m example, should theoretically have more activation to influence lemmas than world the secondary name in the semantic associate example. Understanding the poten tial difference of near-synonyms and semantic associates will further the theoretical understanding of lemma interaction in the lexical system. The influence of mediated priming of TOTs is an intriguing phenomenon, as individuals are seemingly unaware of phonological similarity of the primes and the TOT words. These findings offer another means of understanding spontaneous TOT incidence and resolution. In addition to James and Burke's (2000) explanatio n that spontaneous TOT resolution may occur when the individual hears or sees a word phonol ogically related to the TOT word, spontaneous TOT resolution may occur via mediated priming. Th at is, the individual ma y hear or see a word that semantically activates a word phonologica lly related to the TOT word. When actively attempting to resolve the TOT state, individuals te nd to think of same part of speech alternatives (Burke et al., 1991), which may act ually hinder the resolution incide nce (Abrams et al., in press, Abrams & Rodriguez, 2005). Thus, a beneficial method to induce resolution may merely be to not actively pursue TOT resolution, as it may be achieved via mediated priming. In everyday life, this phenomenon may simply mean that the individual disregards the TOT state, allowing for inadvertent exposure to the TOT words first syllable phonology, whether via direct or mediated priming, may induce "spontaneous" TOT resolution. Given the high number of spontaneous resolutions in everyd ay life, it seems likely that me diated priming is responsible: Mediated priming provides individuals with mo re opportunities for resolution than would be expected if only direct prim ing influences TOT states. 59

PAGE 60

APPENDIX A TARGETS, DOMINANT NAMES OF PRIME PICTURES, SECONDARY NAMES OF PRIME PICTURES, AND UNR ELATED CONTROL PICTURES _________________________________________________________________________ Target Dominant Name Sec ondary Name Unrelated Control of Prime Pict ure of Prime Picture Picture _________________________________________________________________________ acquit fish tank aquarium compass aorta gorilla ape tomato badminton scale balance dice bailiff bikini bathing suit lawn mower ballad purse bag heart biopsy motorcycle bike helicopter bistro mixer beaters elbow boa canoe boat pretzel bogey raft boat disk bookworm hedges bushes rainbow bootee stereo boombox cheerleader botanist chest box grapes bunion rabbit bunny eyebrow cadaver tape cassette door calamine money cash football caviar hat cap star cherub recliner chair typewriter chives baby child tiger chronic alligator cr ocodile rhinoceros chute boot shoe cat compost policeman cop swing covet cabinet cupboard trophy dagger ballerina dancer trampoline discharge plates dishes train dungeon weights dumbbells calendar eligible moose elk sun fluoride tile floor ball garland trashcan garbage can penguin genealogy pants jeans well gingko present gift handcuffs gloat world globe house hemorrhage chicken hen puzzle hostel pig hog rain jester airplane jet mushroom justice pitcher jug angel kaleidoscope mug cup lock kosher jacket coat pumpkin liable lamp light slide 60

PAGE 61

liquidate branch limb kite lotus bread loaf tank lumberjack suitcase luggage arrow lynch chain links hinge mallet rat mouse pipe metronome ruler measuring stick spider misogynist glove mitt fan mount lips mouth queen mutiny donkey mule cactus nocturnal book novel hamburger nomad binder notebook mermaid papyrus nightgown pajamas flower pasteurize pot pan snail patriarchal bucket pail camera patronize glue paste wrench pawn toilet potty skateboard photosynthesis pictures photographs balloon pivot needle pin hammer potpourri mailman postman horseshoe ritual bow ribbon squirrel salvage flip-flops sandals castle satire sub sandwich clock saturate bags sacks cake sherbet blouse shirt tree socialism couch sofa key spay rocket spaceship basket spectrum glasses spectacles feather spritz runner sprinter peacock stock traffic light stop light paperclip stoic rock stone cross streamline road street microscope strobe carriage stroller yo-yo stutter abdomen stomach violin taxidermy license plate tag skunk tenor teepee tent band-aid toga frog toad sewing machine torpedo turtle tortoise piata turban record player turntable fire hydrant typhoon wheel tire harp veto car vehicle leaf warden cane walking stick roller coaster 61

PAGE 62

APPENDIX B EXPERIMENT INSTRUCTIONS Upon arrival to the lab, partic ipants read instructions indi cating that the purpose of the experiment was to study tip-of-the-tongue (TOT ) states. Participants read the following description to understand a TOT state: A TOT occurs when you are unable to retr ieve a word that you are certain you know. When experiencing a TOT, you may know the word's definition, you know how you want to use it in a sentence, and sometimes you can say what letter it st arts with or what it sounds like. A TOT state is different from ju st feeling like you "should" know the word; in a TOT state, you may have the feeling th at the word is just beyond your reach, or you may feel a sense of frustration because of your inability to recall it. If you are uncertain what a TOT state is, please ask the expe rimenter to describe it more fully now. After reading the instructions, a ll participants indicated that th ey understood a TOT state, but the experimenter reiterated the definition provided in the instructions. The instructions stated that we would st udy TOT states by having participants complete two tasks. The first task would be answering definition-like, general knowledge questions that ask participants to produce the wo rd that best fits the definition. After viewing the question, they are to respond in one of four ways: a) "Known" if they know and can provide the answer; b) "Unknown" if they do not know the answer and would not know it if someone told them; c) "TOT" if they know the word, but just cannot seem to access it at that moment, though possibly feeling as though it is "ri ght there"; d) "None of the Above" if they know the word but are not in a TOT state for it; that is, they do not have that feeling that it is "right there", but if someone told them the word, they would know it. Participants read that they should give their response promptly after reading the question, and that the questions were intend ed to be difficult so to not worry if they do not know all of the answers. The instructions described the second task as an unrelated task that involves naming pictures of objects to give them a break from an swering the questions. Participants read that we 62

PAGE 63

were collecting data for another study on names of common objects, so they would see various pictures. They were instructed to type the first name that comes to mind for the object when they see a picture. Additionally, they read that because we are intere sted in their first impression, there is no right or wrong answer, to feel free to gi ve multiple words to name the picture if that is what comes to mind, and to not worry about the spelling. Lastly, they were informed that all pictures were nouns. After participants read the instructions, the experimenter ensured that participants understood the procedure by reiterating the nature of the two tasks and the response options to the questions. When the experimenter felt assure d that participants understood the procedure, participants were allowed to begin the experiment. 63

PAGE 64

LIST OF REFERENCES Abrams, L., & Rodriguez, E. L. (2005). Syntac tic class influences phonol ogical priming of tipof-the-tongue resolution Psychonomic Bulletin & Review, 12, 1018-1023. Abrams, L., Trunk, D. L., & Merrill, L. A. (in press). Why a superman cannot help a tsunami: Activation of grammatical class influences resolution of young and older adults tip-ofthe-tongue states. Psychology and Aging. Abrams, L., White, K.K., & Eitel, S.L. (2003) Isolating phonological components that increase tip-of-the-tongue resolution. Memory & Cognition, 31 1153-1162. Balota, D.A., & Lorch, R.F. (1986). Depth of auto matic spreading activation: Mediated priming effects in pronunciation but not in lexical decision. Journal of Experime ntal Psychology: Learning, Memory, and Cognition, 12 336-345. Brown, R., & McNeill, D. (1966) The "tip of tongue" phenomenon. Journal of Verbal Learning & Verbal Behavior, 5 325-337. Burke, D.M., MacKay, D. G., & James, L.E. (2000). In T.J. Perfect & E.A. Maylor (Eds.), Models of cognitive aging (pp. 204-237). New York: Oxford University Press. Burke, D.M., MacKay, D., Worthley, J.S., & Wade E. (1991). On the tip of the tongue: What causes word finding failures in young and older adults? Journal of Memory & Language, 30, 542-579. Cohen, J. D. (1976). Random means random. Journal of Verbal Learning & Verbal Behavior 15, 261-262. Cutting, J.C., & Ferreira, V.S. (1999). Seman tic and phonological information flow in the production lexicon. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25 318-344. Dell, G.S. (1986). A spreading-activation m odel of retrieval in sentence production. Psychological Review, 93, 283-321. Dell, G.S., & OSeaghdha, P.G. (1991). Mediated and convergent lexical priming in language production: A comment on Levelt et al. (1991). Psychological Review, 98 604-614. Dell, G.S., & OSeaghdha, P.G. (1992). Stages of lexical access in language production. Cognition, 42 287-314. Dell, G.S., Schwartz, M.F., Ma rtin, N., Saffran, E.M., & Gagnon, D.M. (1997). Lexical access in aphasic and nonaphasic speakers. Psychological Review, 104 801-838. 64

PAGE 65

Francis, W.N., & Kucera, H. (1982). Frequency Analysis of English Usage: Lexicon and Grammar Boston, MA: Houghton-Mifflin Company. James, L.E., & Burke, D.M. (2000). Phonological priming effects on word retrieval and tip-ofthe-tongue experiences in young and older adults. Journal of Experime ntal Psychology: Learning, Memory, and Cognition, 26 1378-1391. Jeschniak, J.D., & Schriefers, H. (1998). Discrete serial versus cascaded processing in lexical access in speech production: Further evidence from the coactivation of near-synonyms. Journal of Experimental Psychology : Learning, Memory, and Cognition, 24 1256-1274. Levelt, W.J.M. (1989). Speaking: From intention to articulation Cambridge, MA: MIT Press. MacKay, D. (1987). The organization of perception and ac tion. A theory for language and other cognitive skills New York: Springer. MacKay, D. & Burke, D.M. (1990). Cognition and aging: A theory of new learning and the use of old connections. In T. Hess (Ed.). Aging and cognition: Knowledge organization and utilization (pp. 213-263). Amsterdam: North Holland. Peterson, R.R., & Savoy, P. (1998). Lexical selection and phonological encoding during language production: Evidence for cascaded processing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 24 539-557. Raaijmakers, J. G. W., Schrijnemakers, J. M. C ., & Gremmen, F. (1999). How to deal with "the language-as-fixed-effect fallacy": Common misconceptions and alternative solutions. Journal of Memory & Language, 41 416-426. Snodgrass, J.G., & Vanderwart, M. (1980). A sta ndardized set of 260 pictures: Norms for name agreement, image agreement, familiarity, and visual complexity. Journal of Experimental Psychology: Human Learning and Memory, 6 174-215. Stemberger, J.P. (1985). An interactive activatio n model of language production. In A.W. Ellis (Ed.), Progress in the psychology of language (Vol. 1, pp. 143-186. Hillsdale, NJ: Erlbaum. Vitevitch, M. S., & Sommers, M. S. (2003). The facilitative influence of phonological similarity and neighborhood frequency in speech pr oduction in younger and older adults. Memory and Cognition, 31, 491-504. White, K.K., & Abrams, L. (2002). Does prim ing specific syllables during tip-of-the-tongue states facilitate word re trieval in older adults? Psychology & Aging, 17 226-235. White, K.K., & Abrams, L. (2004). Phonologicall y mediated priming of preexisting and new associations in young and older adults. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30 645-655. 65

PAGE 66

66 BIOGRAPHICAL SKETCH Lisa Merrill was born in M onroe, North Carolina. After graduating from Forest Hills High School, she attended Furman University wh ere she earned her Bachelor of Arts in psychology. Later that same year, she began cond ucting research with Dr. Lise Abrams at University of Florida and will earn he r Master of Science in psychology in 2007.