<%BANNER%>

Covert Aggression

Permanent Link: http://ufdc.ufl.edu/UFE0022335/00001

Material Information

Title: Covert Aggression The Means and Motive of 'Getting Away with It'
Physical Description: 1 online resource (126 p.)
Language: english
Creator: Miller, Marisa
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aggression, covert, gender, indirect, messages, mixed, presentation, relationships, self
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Existing aggression literature covers a variety of aggressive means and motives, but so far little research has examined the nature of covert aggression?aggression that simultaneously contains both a positive and negative interpretation, thereby possibly escaping the negative consequences of such aggression. This study stands as the first experimental investigation of responses to covert aggression and the use of covert aggression. Participants (N = 295) completed a bogus creative intelligence assessment and received false feedback from a partner (actually nonexistent). The evaluative feedback was either positive, negative, or conveyed an ambiguous (mixed) message. Participants then had the opportunity to send a combination of positive, negative, and alternate messages to the evaluator. The alternate messages were either ambiguous, and thereby provided an opportunity to employ covert aggression, or neutral, and thereby provided no opportunity for covert aggression. Participants subsequently reported their perceptions of their partners and emotional state. Pilot participants verified that the feedback in the ambiguous, positive and negative feedback were perceived as containing mixed, positive, and negative interpretations, respectively. In general, participants did not respond negatively to the ambiguous feedback and viewed ambiguous and positive feedback partners in a similar, positive manner. In contrast, participants receiving negative feedback were subtly affected by the availability of mixed messages, indicating that when people can send mixed messages, they do so and are, on the surface at least, nicer than when they must express hostility in an open and clear manner. Negative feedback participants experienced more negative affect and reported less favorable views of their partners than either positive or ambiguous feedback participants. Individual difference measures moderating these effects are discussed.
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 Marisa Miller.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Schlenker, Barry R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022335:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022335/00001

Material Information

Title: Covert Aggression The Means and Motive of 'Getting Away with It'
Physical Description: 1 online resource (126 p.)
Language: english
Creator: Miller, Marisa
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aggression, covert, gender, indirect, messages, mixed, presentation, relationships, self
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Existing aggression literature covers a variety of aggressive means and motives, but so far little research has examined the nature of covert aggression?aggression that simultaneously contains both a positive and negative interpretation, thereby possibly escaping the negative consequences of such aggression. This study stands as the first experimental investigation of responses to covert aggression and the use of covert aggression. Participants (N = 295) completed a bogus creative intelligence assessment and received false feedback from a partner (actually nonexistent). The evaluative feedback was either positive, negative, or conveyed an ambiguous (mixed) message. Participants then had the opportunity to send a combination of positive, negative, and alternate messages to the evaluator. The alternate messages were either ambiguous, and thereby provided an opportunity to employ covert aggression, or neutral, and thereby provided no opportunity for covert aggression. Participants subsequently reported their perceptions of their partners and emotional state. Pilot participants verified that the feedback in the ambiguous, positive and negative feedback were perceived as containing mixed, positive, and negative interpretations, respectively. In general, participants did not respond negatively to the ambiguous feedback and viewed ambiguous and positive feedback partners in a similar, positive manner. In contrast, participants receiving negative feedback were subtly affected by the availability of mixed messages, indicating that when people can send mixed messages, they do so and are, on the surface at least, nicer than when they must express hostility in an open and clear manner. Negative feedback participants experienced more negative affect and reported less favorable views of their partners than either positive or ambiguous feedback participants. Individual difference measures moderating these effects are discussed.
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 Marisa Miller.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Schlenker, Barry R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022335:00001


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 E20101221_AAAABA INGEST_TIME 2010-12-21T15:45:33Z PACKAGE UFE0022335_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 7760 DFID F20101221_AAAWUU ORIGIN DEPOSITOR PATH miller_m_Page_106thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
6e3098e2ebe85a963a82a92f0236847f
SHA-1
5e14353704e6fedcf6bbec8fad21b72e12cd3729
50915 F20101221_AAAVRS miller_m_Page_117.pro
15923fe2fe0fd3a29df7f4dda884b778
583d3038a49ff8994452eb4b05b95b46c3ef3dd6
214 F20101221_AAAVSG miller_m_Page_105.txt
88d6eef1516244a465b8530bc4f0c2ac
8bc4080ad467b353b0e7958741b296dbf0421a37
7484 F20101221_AAAWUV miller_m_Page_116thm.jpg
f823578d24245f86f9131042821aacee
3d0fa5ace213c2cdd99938cf87994288277ddb2e
36812 F20101221_AAAVRT miller_m_Page_114.QC.jpg
462fbaef9741562202df81c1a46d1595
0d21857e7ddcbaf5e196b5046b4ecc57056504e3
386969 F20101221_AAAVSH miller_m.pdf
329efebad323e48075bce8f9668826f9
f87460e677d8e9eedfd8482d4594823c16bd9665
40142 F20101221_AAAVRU miller_m_Page_112.pro
8680053f689a54190fb13ec5d8d9c9c1
d0668812aa14f3f0eb89f9d394b4cced79049d3e
1051951 F20101221_AAAVSI miller_m_Page_039.jp2
47672f928e249cf9ff8ad02b2ec74570
442be22285872c36a83bbc25b13e25eb39a2ea1c
9100 F20101221_AAAVRV miller_m_Page_122thm.jpg
d215a9a71ab6ba2bd6284c1f0174a195
ba89ad0fdcb1494cd6bb58c4fbc1d78828e2c722
560951 F20101221_AAAVSJ miller_m_Page_121.jp2
a4374f5ee1345007b3853a287df343ff
8a761de4eaefd810f73d8538ae2108fc27b30cc1
8696 F20101221_AAAVRW miller_m_Page_124thm.jpg
cae7d4d6136affef27dce38b53adb5a7
33bfa744de71040be697285ace1e8ed3a988443b
25271604 F20101221_AAAVSK miller_m_Page_077.tif
1555230b6be0f6317407c9b5f7c59894
dcaa72aa4eb8cc1f1581e84298fed68727202898
112313 F20101221_AAAVRX miller_m_Page_035.jpg
3211634a92a20e8db77af32ecf6d369e
f8cba356ec2336fb14eee106878ac3e734eb5679
35342 F20101221_AAAVTA miller_m_Page_029.QC.jpg
999d3a280e6979c18964d35566a4a395
0afb9e6ff4b6991913b0ddd27d6471804f0fd7cc
927818 F20101221_AAAVSL miller_m_Page_117.jp2
be9e39c87f5675919ca235294d5eca44
f6b3a48f98e24fd6261b3760818b50846a618e0f
20736 F20101221_AAAVRY miller_m_Page_074.QC.jpg
766a0ecf1e4bfa63be3755a7d0e6ccd2
2803f60a0bc7ba6fff7e5ece9b894d164b4171c3
54604 F20101221_AAAVTB miller_m_Page_066.pro
803d724889a553f303a8d5972b0aed32
06f00f1edda79e0a59f4265b503a375ddeebe5d0
112959 F20101221_AAAVSM miller_m_Page_059.jpg
b29f5e4ec18f2ca743b209e37a01b7f4
a739f71ec94234a8e1b2f604f53c5f97b12f8ab9
109419 F20101221_AAAVRZ miller_m_Page_099.jpg
6d621a902489fdb3232a7a0284689be9
43f9ee7c3464e8967578c9f74d81a1465a588baa
F20101221_AAAVSN miller_m_Page_105.tif
ccc2be16a35040ab94e9461155178ca6
40d3cdf2e200ee75fbc7590c34787bc620af7444
1051980 F20101221_AAAVTC miller_m_Page_036.jp2
497211f85c45054964a929405614ecab
aa23a317fb1cd7bdc5e88dfd565c17474a3f863a
8878 F20101221_AAAVSO miller_m_Page_091thm.jpg
bc504f06d4519de860bd6ae7e27fc560
3649312342280b92bdc6317df7a618b1dc8ff846
2139 F20101221_AAAVTD miller_m_Page_088.txt
a7e155f6651103571c89d0a6266a9b25
e2ce20c525d45cc5da4f4d5d0d97075e4ff7af10
1181 F20101221_AAAVSP miller_m_Page_002.QC.jpg
e51d79158e18b47ac4abb97c4e2f6b1b
913df9f28c7f9ff8ac8b60fbaacb3c45d739dc62
8971 F20101221_AAAVTE miller_m_Page_035thm.jpg
4f03ece52b0444436e297905e7899cba
a861bf84337abef153be5dd76fcd0ff57c73509d
9900 F20101221_AAAVSQ miller_m_Page_086.QC.jpg
860d730746581b4de9da516a6bae5b78
565e71df442e7171df4d637865fe3223752c0bf6
38139 F20101221_AAAVTF miller_m_Page_119.jpg
5108feabb5ca18ec83504b5629dc1588
7de032b18db278369932a965225d9e7b62f2e636
35854 F20101221_AAAVSR miller_m_Page_035.QC.jpg
0b89b2951159bd228c1fe603f8104039
9cc9a8f68c34fd6c1b6057bd9143aa525ea5223e
875408 F20101221_AAAVTG miller_m_Page_112.jp2
5bed7940c4bb1b0cd8f8cd4c9fcebf45
1cf0ed4e1fffb2c585d404ef9eee2e4cb64e14f6
24553 F20101221_AAAVSS miller_m_Page_077.pro
78793aa986f704bf2d20772681b12bb1
9b682f7831dd3ebb98c6b4d6bd38a971a93520f7
F20101221_AAAVTH miller_m_Page_068.tif
8bd22465781c824e08f6358cb18089a3
ca13c5af6f1dbe929ff8841b4643837d29f841a8
86126 F20101221_AAAVST miller_m_Page_117.jpg
22e051bcc4d090a77141e3b16a20a9d7
79f0d4fbbf379498e503a791d601a01a82ba781e
1051956 F20101221_AAAVTI miller_m_Page_059.jp2
3f51affbd0e95e1c9ed42dcfb7237331
7fa7c4c82ee503cf3937ab90acd59a8297bd3b39
53463 F20101221_AAAVSU miller_m_Page_060.pro
991e60a617ac83e01c1811758d8050f7
598f727cdb07630d3a04b777c78de323a73c7506
33336 F20101221_AAAVTJ miller_m_Page_008.jpg
5c389f78de18be0472b44c32408a873c
2ec7491fbca2600ad377ff481892ea38b79ec313
34891 F20101221_AAAVSV miller_m_Page_064.QC.jpg
7963b177810b3081cb089a0023c6d10d
6893ff142d666dd1482fa86c607523888a752dce
108513 F20101221_AAAVTK miller_m_Page_054.jpg
75123dbb612a3be4daf7cce781bd47e8
7011bea531e1f1c5af5f7460b4aa66a110148aa1
2143 F20101221_AAAVSW miller_m_Page_025.txt
b3009f81351b18e3ce9d70dda7d234d2
9cc7b9b713f7b556b7942007f075dba0cd1a5665
1051971 F20101221_AAAVUA miller_m_Page_065.jp2
6f67c92db6947d16b528e6776ac1e94a
395b0a35abe20914ff051bf0fd569f493c00bf91
2178 F20101221_AAAVTL miller_m_Page_014.txt
e4110272c3110869b587af3e7d7d0094
f726c5726bf87cd5b3290363c83c82ed024eee8c
36562 F20101221_AAAVSX miller_m_Page_040.QC.jpg
9a2ebb52cf76eab5543600bbad2bf2aa
a786bc25331ffb74c7ea75db1f816db293ea3e6e
8808 F20101221_AAAVUB miller_m_Page_017thm.jpg
2f9ebba646b9d700f3e532fe625e79ce
4b7ab29927853649732c68163721887c25173d18
F20101221_AAAVTM miller_m_Page_052.tif
d157ef8df2dc6443dad60ab2cec2ccdb
e83288fa619ecc75241d9fbfb387e3407532e1ef
F20101221_AAAVSY miller_m_Page_108.tif
13bcf1aeb5ee801d85951048510b7d38
ce561ab55fef50b76aac0e50ada09cca41eeda46
F20101221_AAAVUC miller_m_Page_092.tif
1b515f9495ff2dee66409a4c4a49b4fe
6f7d694d792d49b40e7d48ebc6a6d4b80db71a3b
8939 F20101221_AAAVTN miller_m_Page_088thm.jpg
411072b9e5a5eb6aab119ce9fc51244f
cb663672ef4c46b5f8c49817d673101786c5f1fa
1051972 F20101221_AAAVSZ miller_m_Page_067.jp2
1130fbe54ae5f5d278e033209eceacd7
37251f3173283e63641d169d277bb59941fa3311
106808 F20101221_AAAVTO miller_m_Page_064.jpg
948fee4424d9ba59c1543de3eb64e4ac
428bcdd66c24bd513ec0f2cf995091412af14553
2125 F20101221_AAAVUD miller_m_Page_064.txt
989789b7acbc20d1633d8156e2269579
b9b768591d429c5878888787ec17b72860cd76fe
25265604 F20101221_AAAVTP miller_m_Page_080.tif
b85e09dabe441baf71a57d5b47966df3
f3e6ed36ef248b9f38ea4fb2dc742b2642ecb6a7
17843 F20101221_AAAVUE miller_m_Page_048.QC.jpg
c963150236ed393b4c2fa4002d1b8a0b
5fe04b34d996ddc3b6af77ee80583ada8cf4b434
37487 F20101221_AAAVTQ miller_m_Page_051.QC.jpg
51a8598857861d8666ed393a649eeb7e
f911f91adb51f5a54275dfb10ede259309c5119d
111429 F20101221_AAAVUF miller_m_Page_034.jpg
6046049668cd32a38348a0cdd61d67c5
78f06c2cf0e4efa3dfb6b99e761fd969918ff489
F20101221_AAAVTR miller_m_Page_063.tif
ac25987a646fcd6a9d28fd622bfae5cc
e93144e3be8f0478dcda63a06b1b2a3aefd80a55
F20101221_AAAWAA miller_m_Page_016.tif
e4243d90e80cea1355202e27d0e951ef
c7ced96532bbae92e9142aef439e01440b469400
49640 F20101221_AAAVUG miller_m_Page_116.pro
ce1c159493ebd572fb7cc57c44e9ea73
b8a40abe8a3cbf1f589200ed24f2ae3018fcc7ab
55981 F20101221_AAAVTS miller_m_Page_034.pro
07017af0242810bb8d70e39c6f0546cb
186dc736c83eb958827c0fb297b89c8179d9b186
2152 F20101221_AAAWAB miller_m_Page_047.txt
d678e4d41c63d29d0379b25f4fd127c6
c832140ac60f67298a1b98fd3f250eeb5d7bdee3
1051986 F20101221_AAAVUH miller_m_Page_062.jp2
fd96596a246753af4d4b1b1e9ef8caab
3c1a08a93cd0423c174aeb63db8691afd88e3aa4
F20101221_AAAVTT miller_m_Page_112.tif
a2110ab74f7affbed347097f905b5c48
76f6a4c55d14852a624be70a0ed849381c395ab0
3345 F20101221_AAAWAC miller_m_Page_086thm.jpg
bbf277b24a83aae00a2111d4275a1cae
d288b85b6579778d6cc2ffaad850ed8d0c363d7f
52910 F20101221_AAAVUI miller_m_Page_095.pro
0a000f64d69df8fe0f02999cb5174d4d
6209e8bd2b9246eb8587779d60f00a19fa0883ed
109652 F20101221_AAAVTU miller_m_Page_088.jpg
53a7bd11306bf987fc1af857203b8b30
223cde0c03193dd4ddc4c806cf67ac3e639e5592
20857 F20101221_AAAWAD miller_m_Page_119.pro
c91c584f9f51d505d254c19471321f0b
87902d379ec98c8b82b400ce0276b3bc33f38dc3
F20101221_AAAVUJ miller_m_Page_036.tif
40eea19f81721bd651265e4ab45e14f6
c3d76a7b12f463a5d733a3df96b289dfb7afe52c
46568 F20101221_AAAVTV miller_m_Page_078.jpg
96539aec8f052c8de4eb2ff7a5eff453
8522ee771260a11914fec7911a62484b6d2e2ade
57857 F20101221_AAAWAE miller_m_Page_024.pro
477f3122e32be543d041700f46f689c8
673b54d35cbeb9e7a86fb50f1a3d67c2de1b43c4
50592 F20101221_AAAVUK miller_m_Page_107.pro
f50ec2837df9c993ec03150e581974c6
d77f120c7228b679e7bb669f87f189195bb4634e
7968 F20101221_AAAVTW miller_m_Page_068thm.jpg
a05e3ac7b8006afd4c92e01f0eff4365
095ca4cbd233f739a7845ec8783d03667edfb4ab
8768 F20101221_AAAWAF miller_m_Page_040thm.jpg
1a4194c4b8d7a59cc15ab950b8fc0134
fa7cdf38ee8f8a0e145ac965661d288206263aa6
22979 F20101221_AAAVUL miller_m_Page_112.QC.jpg
112937f9e327a47c9946208117edad82
4fd49972078ce6e5ee60e0474eef016194aba15b
23287 F20101221_AAAVTX miller_m_Page_121.pro
0186da6428fa4fb6bb944114f829d450
c282e112fecb98c3a382d0a0524ef8fb3c37d2f9
9176 F20101221_AAAWAG miller_m_Page_093thm.jpg
018a0fd863a09a32de7057a8c236260b
e879f8bfc904ea7e4b19377daff2117df00723e0
98379 F20101221_AAAVVA miller_m_Page_068.jpg
3b3a5cc65dcd20d6480b209346271756
57b297977fcfe8494cf87c987ed2ad0d7a83bae1
35392 F20101221_AAAVUM miller_m_Page_124.QC.jpg
aefef3ccb8b4870fa8d1b77135d7f38f
12fd624912b6721b6b28eb693d5c23c579d4e9e5
2079 F20101221_AAAVTY miller_m_Page_095.txt
3391fc0f75ef0f4b77b806fd1a0b0ea1
fde4056544e0f97ccd88f33c5e7b174b299baed9
581536 F20101221_AAAWAH miller_m_Page_120.jp2
25bee69e8e7e423b54dbcb494995fd93
0064cb13d99abe0e8f5c5cea1ac5a7a9e7aa98ff
6589 F20101221_AAAVVB miller_m_Page_083thm.jpg
f41af3a3a6d779ff2d6e27b92723fb80
fe781c9dc96d9ed9efdb02387f5882f4529ee88c
9167 F20101221_AAAVUN miller_m_Page_103thm.jpg
de7644a54da810ee41c8690009d6f490
ebb2752ebb6356bfb321546922a769d1ce08f440
114110 F20101221_AAAVTZ miller_m_Page_005.jpg
63b9301965427f28520e2ebc785211c7
0c35c0e9c45e9d22f2b32346f46ee0ee662e4bd2
98478 F20101221_AAAWAI miller_m_Page_049.jpg
560f5b19f170f068f987dd5d54d9f28a
8f2ef12c7336ab36e1339c6e39da45ae9d9feb4c
112873 F20101221_AAAVVC miller_m_Page_058.jpg
d3ba893e7a7723be8f9f3147cfe57111
2e2738ac37afebe454dd338e1d06b6757a320c97
7641 F20101221_AAAVUO miller_m_Page_009thm.jpg
12eaa03e6aeab27362a8f23eca666953
65b08a865978a3f068fd0b4c4b18fe2cdd4255b3
8918 F20101221_AAAWAJ miller_m_Page_071thm.jpg
c4ef09f899b0b149daf15093218db828
2ed47a1bb1547fad9a507e5608e9a6feddaeb847
2194 F20101221_AAAVVD miller_m_Page_059.txt
b8649a11a3bc4ba0c0673412afb4791d
7e244da7dbd2f88fd38781c4c0645e400fd3bab5
F20101221_AAAVUP miller_m_Page_086.tif
8abbab4478360b5cbb7d5549f7acc721
fe944eafe2e282c225d542592337c1f4bd5d67b3
F20101221_AAAWAK miller_m_Page_116.tif
ff6c1f13f53aaa35754e01da698632b6
07cd22e092ced25bbe84c329cc32af22ec8880c5
8760 F20101221_AAAVUQ miller_m_Page_054thm.jpg
ef870fb17132c1bb2e257565637d0223
1fad306dec708c2e5046e381fcb580b6b34a1802
9147 F20101221_AAAWAL miller_m_Page_063thm.jpg
b95c20090a2400d3fba8365ef178dfb9
7ab0bae205f706a4a6dcf3b5a4029aa10155a822
25355 F20101221_AAAVVE miller_m_Page_107.QC.jpg
7e84b737d5aa93b3142dcc24343589ab
8f9c767dfd47383d3c56f7539d0d25899f74a49c
1051984 F20101221_AAAVUR miller_m_Page_104.jp2
f0e23d4672682fd640f0237a659d7437
fb60379b44da5655284d55c0e0fca66baaf30b6b
3892 F20101221_AAAWBA miller_m_Page_002.jpg
c44d0374eb34359f2c8ca33b58a700ea
31f432738ccc6c738a2b4c65e25ab3e5073094df
97249 F20101221_AAAWAM miller_m_Page_042.jpg
e8cff6abc3f887f985d983589112df28
7849e73108bbd7ea3c260d9823cd61eedaf2f136
34181 F20101221_AAAVVF miller_m_Page_095.QC.jpg
b2445cdefc0656a1a44559df3eb6991a
16f56e3a7b443d1af581e6fafca75eb2e9c8ac28
F20101221_AAAVUS miller_m_Page_117.tif
fa36097d3d413369a203f0c71adc4aeb
4c597107bdf8ad1f37e128e7f9e208577d15d3a0
404059 F20101221_AAAWBB miller_m_Page_086.jp2
0f1131cbadfbf04d79939155d5dd10f5
35fedff1fea6a13148e427c619afe4f5f4a89d4a
F20101221_AAAWAN miller_m_Page_040.tif
3b812675d74ceff5e08c3a5151ac9307
62902175fad97fb29c55e0436d3e622c36f2c150
37320 F20101221_AAAVVG miller_m_Page_046.QC.jpg
90a46797df2df573e9ebc8f5050493bb
8821561f1e2424b37065e9de59835a235e323ea8
F20101221_AAAVUT miller_m_Page_007.tif
36649c85e545ac8c5ab3762f979d0a0a
a4bc521e6bb19a625d8c7c308e2991d641140c91
74037 F20101221_AAAWBC miller_m_Page_114.pro
a84a71c3b74de23015950d955ce20d73
e58786157d30a6cf8d563d9519e3cda65f5dd531
111865 F20101221_AAAVVH miller_m_Page_098.jpg
8e020d3a36d9809de27d794a74a47429
460716801daf71fb3813930e07b769e7c3dce891
124058 F20101221_AAAVUU miller_m_Page_073.jpg
06e42e0f5e4d232b9f35c5f95a96aa68
e56df48da7f8f89bc08c8f0532d92a515f42db1e
882 F20101221_AAAWBD miller_m_Page_002.pro
6607a18118013ee41e7607ad11f5ffb2
d8808b988ca7405116fa7a5663c411ef3d62de3d
1051899 F20101221_AAAWAO miller_m_Page_035.jp2
11df03a830642da5a92cdc21a912eda9
bfc66ca3f2838908d1e7e5d97b85415ff22183ac
929 F20101221_AAAVVI miller_m_Page_078.txt
8d0f44fbf13430e80d6e5155526f1f33
4bb4d7d60f184056e356ebcd84213e3970b6d703
F20101221_AAAWBE miller_m_Page_123.tif
33022b0fd30307502aef85f7fad9ca94
e4676b7ec4f5fd4c9abd872d0aaee5669dd2b6cf
54740 F20101221_AAAWAP miller_m_Page_090.pro
483e125d9efeaa3a237e84a99ac1f76b
b12a02649c2d0f7252a40ac087d23b5bf8731bec
25559 F20101221_AAAVVJ miller_m_Page_076.pro
6107a5cf68965ebc45c6f3269dcf6ee5
0e963557c5de5a5228c4c32c6edee899c347d654
1051943 F20101221_AAAVUV miller_m_Page_097.jp2
bc1ae25f0a4b477d828eb82158f6fc67
698eddc898cbd022b1269433f543a9038c0911e6
100261 F20101221_AAAWBF miller_m_Page_009.jpg
e6e713f6a991332deb96938d5d12168b
2225aeaf8ee75fe481043e2ea50c778ff0c6adac
2204 F20101221_AAAWAQ miller_m_Page_072.txt
28423ff14de30a493b7dcc8abd0d4c63
6a75ab0efabc0a4500ab970ad9f35876f377c955
36911 F20101221_AAAVVK miller_m_Page_093.QC.jpg
6617f8bb2cb3c61ad9607f0ac7142730
e9047166da92931ead2669ca06f19e359305390f
36316 F20101221_AAAVUW miller_m_Page_113.QC.jpg
5d432ce2f4feae84f48105fcd6e46a19
a6c81f01667c7bc4abd26501b25845cc4b030653
F20101221_AAAWBG miller_m_Page_070.tif
0af3b50269b76dff29dea9e72cc03489
eeb8c15f5b3dfa236886624bb750f8bd147eeb3a
232174 F20101221_AAAVWA miller_m_Page_001.jp2
db82d6451642c3edfee66da0080e10ee
0bf874a621721c3a0831f79f3c569aa40995d733
27964 F20101221_AAAWAR miller_m_Page_084.jpg
fcc1e47eb804fde31dd5db7bbc7d3fe4
05cb8aff9f8d4557d21e2cc3958be1e872cd1c44
F20101221_AAAVVL miller_m_Page_024.tif
6be772316da9567639519874f9e00418
ab9707aab4d0cc40adfe80b6a5500fd68c5112d2
11936 F20101221_AAAVUX miller_m_Page_010.pro
6ebcdd4161c6a7b7f3d157291935b566
1fe693a67c106eb95e455f46047440f052adfe67
38065 F20101221_AAAWBH miller_m_Page_024.QC.jpg
dd74e2e05f5ea694929b226f8f415d3e
9b6f3798ea5242eb12f68a8a3bf81c031207b540
146396 F20101221_AAAVWB miller_m_Page_037.jpg
bc05212f912cdd582e1111d4fab5b30b
9a7d520a2c9abce2fa3fedf0c8867a75e18787c0
1051982 F20101221_AAAWAS miller_m_Page_024.jp2
afcf88d3c5c0072413f1e5fdee1fd45d
26498911ec729cb6d781d3d12322c4177b260a87
2149 F20101221_AAAVVM miller_m_Page_090.txt
67153a8fa15c9d1a3551593258540006
b19a4039ac6aa85c5d467a1c911152d12cc421b9
276953 F20101221_AAAVUY miller_m_Page_010.jp2
b56459f574ff60302a7730717dfdaab4
12e4d6c1546f91e79c629a3e07459484fb69a1fb
38769 F20101221_AAAWBI miller_m_Page_096.QC.jpg
584b1852636c31b517baa187534ba26c
58778d97456d7a35095548b176641faac8fcfdca
F20101221_AAAVWC miller_m_Page_018.jp2
f898e31f876316b36005efb655938bd8
ec44f9c6438434de39f5a049d570be2691576ac9
F20101221_AAAWAT miller_m_Page_001.tif
a31335e87b12c71e01d633ecd91c46f9
8feb6c1b625ae71972a1620a6fcd5b7216477f01
3297 F20101221_AAAVVN miller_m_Page_114.txt
069cec9ef25abbab019926430b00f602
da0d95d49a385c79f46b2ac8400f38f84af337c1
2132 F20101221_AAAVUZ miller_m_Page_054.txt
055019e8377d5ed4a8bc0bad19c4a6a2
709e9060be5bf766a15c6efdb2f4c6ffa4e609d9
1051954 F20101221_AAAWBJ miller_m_Page_057.jp2
50f2b74b31d444b120cb0d1e42a3ea32
5b119e6f2b1f3b211a9ec266a6b6afacd0562d8a
56365 F20101221_AAAVWD miller_m_Page_069.pro
33eebf432689b14d29b0a11f97760ba9
a1306c9434f21e619db4f99fc02c0407826d6bf6
8955 F20101221_AAAWAU miller_m_Page_097thm.jpg
cf540bbed034386f00d935ac98f334cc
c64624bcd820d4ba73fc43ab0b7aca376721c338
116858 F20101221_AAAVVO miller_m_Page_036.jpg
aefdfb98ee6b1a4ad94113c40cd30eb4
91d9fb160812f6352b0f891658d215b2e2fbba34
8294 F20101221_AAAWBK miller_m_Page_114thm.jpg
9b775768169f62ba449c5328e3d5022e
b7717eaf503d6bf1b742fd89f8a6e08fc2553ef6
1051923 F20101221_AAAVWE miller_m_Page_038.jp2
69a0cc160c6eb6f52b8e590895ccb350
98a0bf306306a5be92b4b6ae15b5b0c432d23f3f
39095 F20101221_AAAWAV miller_m_Page_080.pro
4bc4c359cdec32eca31f0d4f57f16d9b
eac0290329fb5ec32bc51de5d5164a353ebb1730
115364 F20101221_AAAVVP miller_m_Page_063.jpg
6619e9f3f567072de2333d7dafcfda3f
2babdaac2d3d5215b8b0581764ad895ba4a01f95
4180 F20101221_AAAWBL miller_m_Page_076thm.jpg
ac0181e4a1276c9e073c341c84b865d3
a8dd810a1f158c8cb8e0c1ae278bd2cf0abd35c8
52066 F20101221_AAAWAW miller_m_Page_077.jpg
2600a927091a7513ef4887ec1e00de0c
411ab2f28289d4351829cabf073efba15a2f136a
34859 F20101221_AAAVVQ miller_m_Page_025.QC.jpg
2add17e0448d55794b671ddd8e6c63a4
6dc3d8ddf1cb4825d1c1fb999a7af3a87569c25c
1455 F20101221_AAAWBM miller_m_Page_007.txt
091d7bd1362aef953dfbace50110d101
3a559bafbfc6d028810a74708e0eeaaeab33824b
1387 F20101221_AAAVWF miller_m_Page_108.txt
c76bd5e332584eaeed51cd59137c8c8b
ba040f66b19754087e045924bc939ee997aea9a7
9952 F20101221_AAAWAX miller_m_Page_081.QC.jpg
2a90a62ba1cee1453aa85dbaff58fc65
3334f40209bcf31b432bc0bf7012c9c883b1c2c3
15467 F20101221_AAAVVR miller_m_Page_076.QC.jpg
0c97061c590ab2a221a36accee79aa9b
f6e5c6e37d59dfe27cd3bcdb815070b6f64420c1
F20101221_AAAWCA miller_m_Page_023.tif
24a0cdd5eb6eb2c4c75aba8b9f1cb4ec
da8c47bd4b23bcce5d43a09cd378cc1468073d96
49277 F20101221_AAAWBN miller_m_Page_080.jpg
0cc86aab507408c14cbc4d6ef5af631d
cd31eef510f8afc678c8b4e21ba63b161fa51daf
3098 F20101221_AAAVWG miller_m_Page_085thm.jpg
d69082468f1fb2e7586462f0cebd049c
5d1fdfb7201c936be31ed08aed0d5f4e7ab70650
4295 F20101221_AAAWAY miller_m_Page_115thm.jpg
64dec0c78c1fbfbbe9dc4a71b709c20b
c3e206d6622bff4c4bec18f9ce2496f327c49aaf
8269 F20101221_AAAVVS miller_m_Page_018thm.jpg
81623212e38a3031887644f4e100c340
a82612b56f4827e9875429b384f54fd094fd3cfe
56755 F20101221_AAAWCB miller_m_Page_056.pro
f9d003793f1b20aa4d24a43c27221883
b0e5727480d505d5d6486d0ac130aeca0d986dde
8964 F20101221_AAAWBO miller_m_Page_030thm.jpg
62f68113e6e3a6677b8192f04b0d6479
e6fc1126467d3bd6e09fae5639436b6112ba849b
66725 F20101221_AAAVWH miller_m_Page_007.jpg
8c6776a40884602ca053c52742e2e356
99e3b19f34c0e5287f756527abcfb6ce1ad9cfa7
54356 F20101221_AAAWAZ miller_m_Page_100.pro
7cb878ec40d765cac2f3ac1466c5e181
41bf54a7e303b5f3e9cebf7c399f20141d116b4d
F20101221_AAAVVT miller_m_Page_101.tif
3205ebfcd4ad24408c3a3da469159a3a
760d26fc3319888fb8ee620128182a010e72504e
F20101221_AAAWCC miller_m_Page_118.tif
3329040bc4301bb735ad55119d1c68f6
78ed8c6958695529c91311797b40019c64eb03f5
113456 F20101221_AAAVWI miller_m_Page_033.jpg
db8da1729ab20509ce607d82817a35bf
e4ddbaafd682c0af2d5b22f2b8ef6cfe33f4d87d
8862 F20101221_AAAVVU miller_m_Page_062thm.jpg
7fd34375c83f523626d5a4f2113ed0c3
c635f5007c2fe21af8e57ecbb446d0487d12acf2
F20101221_AAAWCD miller_m_Page_011.tif
aa610f70e1f098f98462196e78918b36
508499ec417aecc102fae7eb5683065fba577fe6
111301 F20101221_AAAWBP miller_m_Page_091.jpg
49f7b62b54a43b1ac36b04ae3f509541
65db91f828040052c9d0adb732d992cff5e9b3d6
54415 F20101221_AAAVWJ miller_m_Page_088.pro
a41eb6a97fac1a590d72e46c45f90bb1
869395e7e0707a7049bb1716ba7cc148376488fb
8362 F20101221_AAAVVV miller_m_Page_064thm.jpg
c4605e889462035b2af6ddca4b2d49e9
6be12047bfbbde060247ff29ed2c7c19c4c8e9cd
2299 F20101221_AAAWCE miller_m_Page_013.txt
48e99f94cd32a2c773f5e43040ada78e
eef951e18774698fc0c7b63cacbb1ec29f0fbe5d
56044 F20101221_AAAWBQ miller_m_Page_059.pro
e732f0aeae2dbca25479378d8464213f
4744f2f8ea961e8c15c2ead6730d4af489fb9ce3
14726 F20101221_AAAVWK miller_m_Page_126.pro
bb1298df6af32858c49548f8b23da6a8
4053bc4b7231c18a9e5fa32a2bd3c3d31eba440a
F20101221_AAAVVW miller_m_Page_025.jp2
d7e6f9a753f2e78432d8808b1013d1bc
31e02fabb57d970a8bc1243b713c36a0cc387aa5
F20101221_AAAWCF miller_m_Page_043.tif
c879b537227248a32a0f2d1445c646b8
acd2d76d52a8d10f43eb549f6add54f5484a0b13
112173 F20101221_AAAWBR miller_m_Page_055.jpg
0c101fa9cc3dbf9b0ce5ee9841bcdd53
9e39b833439f8bcb2bf509f407c3ffde9a710642
1436 F20101221_AAAVWL miller_m_Page_109.txt
aede0ed73c22839928e492b181d65f14
2abe48331301ecf5a7e241f8df64b1962504e4c0
8835 F20101221_AAAVVX miller_m_Page_065thm.jpg
71e12e21e616c41f97645199e3b5d983
4981aa2bd4914c8b86457a45235fcff7755585d7
35445 F20101221_AAAWCG miller_m_Page_126.jpg
164f5b9062cbaf7699cddf90cd000aa0
1db37e2aef13a8d3bdbdff3a3b59e895415cd32e
8503 F20101221_AAAVXA miller_m_Page_032thm.jpg
f208e35fe0b8c5d7a4a5e8f20370e1b6
4072b858da33fa5acd9803f5fb1d547c9ddc7a31
7352 F20101221_AAAVWM miller_m_Page_117thm.jpg
1b460d6671761d2cb336802573324ceb
cd28e6c6f2b757ac3f9ad737ad3f6820eec8cdbd
806946 F20101221_AAAVVY miller_m_Page_080.jp2
554e73b8f1e74ca599ed92ba848fac36
df489a4f0c6f3483bfff913e2ad447828b4025fd
64438 F20101221_AAAWCH miller_m_Page_122.pro
e78c9f938e4f056b4addadf2ed3577e8
227a2ebe469860c0c2fdc6da0f0de04677a523b9
36188 F20101221_AAAVXB miller_m_Page_109.pro
c3990f0950fae8c051ea73bbd4ebc088
9195aa8a12c7cd9508e234f7ba4ec6134435a228
48133 F20101221_AAAWBS miller_m_Page_115.jpg
01b7b298ca8e7db1224dc13d5a8dd5d8
dd8bcf8e9d10875299c8cd09646dc3187753d971
3423 F20101221_AAAVWN miller_m_Page_082thm.jpg
05992ead8572cd356f08de1ab9f91d96
b68c3af6e39d3b4e694360d8d6b2113903562d3f
F20101221_AAAVVZ miller_m_Page_013.tif
79170e67bdc4d63de8a9a7db0bdecd2a
21c6a196fd2bd2b49c37068fc83ce9c5ffd84d94
7900 F20101221_AAAWCI miller_m_Page_001.pro
82f218a09a7204181bb9d36f736d4679
a9dc2868f9a1bde86dcc110a9be32f086c768167
2836 F20101221_AAAVXC miller_m_Page_126thm.jpg
88062835ed5e31caec4ca55c05f4f119
4be34a637da651a84b41432379fc087067aa78ec
4071 F20101221_AAAWBT miller_m_Page_079thm.jpg
0ab433aa682805d2e9456e9ff3f1bd29
e3c528cd2d034365a071830fa49a1bc199813523
2225 F20101221_AAAVWO miller_m_Page_044.txt
8bc4634535d7cb9612c9f1e9b6584773
70897a9aefd6d18ebbb39941027c1ca66bea522f
36635 F20101221_AAAWCJ miller_m_Page_102.QC.jpg
94be1cc1f5a5bc0bdead1f6cd0663f2a
fb314f2604a165641ab52ee6351517828be97f46
53798 F20101221_AAAVXD miller_m_Page_022.pro
800dcb9017211e5ed7823d223345818d
199b483e5edfcd87bf2cb6ecd0633aa4ad3353d1
F20101221_AAAWBU miller_m_Page_048.tif
a6e31ae8a301ccdb78171795b9fb33f8
bb34c95395dba4c5ea2678d61506c477606938c1
8195 F20101221_AAAVWP miller_m_Page_070thm.jpg
c17e915b284c6237d1371f5d2a12eef3
309fe9f8882303ef7975064e00816614fce61797
8684 F20101221_AAAWCK miller_m_Page_100thm.jpg
0ada3e385c54a9858e671b88bea72bc4
2f3dfe0ea8b5ae957488fae817b25a35aa0a8047
108769 F20101221_AAAVXE miller_m_Page_026.jpg
3f54208819b0e728d18a44ec01c5d746
07dbc80c93ddf10c1b9616d72f56e005935adc75
56883 F20101221_AAAWBV miller_m_Page_051.pro
82d6d422b3be3b1f43f4c5dcd43f25c7
28a8d07bf67b90a997ad2434bba710f5c4e82eb9
21534 F20101221_AAAVWQ miller_m_Page_078.pro
c720f02de6915ce79a5fbc9cf56f0f07
0102e7246e3febc2acabfe14e694436431fc4918
2244 F20101221_AAAWCL miller_m_Page_046.txt
51529816bceba86eacf11af72da1a5c6
be3cea382b146c6129731883933e3f521a0d8510
40934 F20101221_AAAVXF miller_m_Page_037.QC.jpg
13ccaa427868d87f8557778e131a937f
505fb243ff67e89efff940eca1f458878283fff4
56729 F20101221_AAAWBW miller_m_Page_044.pro
27d366d7bd189277273af40e5cfd10b5
a0f5ab553a7fb2fb6893340fcb4c6fe4bb3371a9
1051960 F20101221_AAAVWR miller_m_Page_095.jp2
2e993044cd04e7335a162349c2bef9c9
1487e4cf92b3f155315bc738689ce5cbecca60b2
38506 F20101221_AAAWDA miller_m_Page_056.QC.jpg
f59367d2a0378a431f32c8bebab5a36b
b4a03083184a1618d82033abdaaa7d8ecd1bd06d
1051952 F20101221_AAAWCM miller_m_Page_047.jp2
a2437f3d84878a2977abb6ed011dd3e5
cfa5e53ed8452a5840790f4086cb7d5e54935dc4
8726 F20101221_AAAWBX miller_m_Page_014thm.jpg
5f1221fa261408dd54952b56054b60f0
db20bb59351ed1bdf34958c73a18502226d1f615
2226 F20101221_AAAVWS miller_m_Page_021.txt
fefae45af095a332bf3c09283ffc86f5
eb178e2df576ebb66f472982f9fc3783f8cd5b6c
112752 F20101221_AAAWDB miller_m_Page_094.jpg
4e0e23f98abc722b7faf5c919d24a9b7
c895f6e8ab26587d3d949c36f2e1e653d3fb407d
98 F20101221_AAAWCN miller_m_Page_002.txt
3e75de3f062fa83870e362689967fbb8
df544c62807af3cfbc98a0fb22db62d228b21b40
13253 F20101221_AAAVXG miller_m_Page_008.pro
446f9a597d55bd9ad3d06a2a2701776a
08f375313ff6e3e21200f4f8101b10d5ea901f15
1913 F20101221_AAAWBY miller_m_Page_001thm.jpg
8cfc422bfe59daf7819d62ef84e13219
d6836c7e4c32c8cc344e8d954f513ee31bcb3ba8
1051922 F20101221_AAAVWT miller_m_Page_033.jp2
20787e6775388f60d084736b71fb2353
c19204f80cc9256eeab6abd2041501f1531b6514
35396 F20101221_AAAWDC miller_m_Page_030.QC.jpg
8ee449f2296c59a566420c14d046aec6
338ff366d574be6facfd23759f134262b32f38a9
16750 F20101221_AAAWCO miller_m_Page_075.pro
7d11474486535d2ce8fbfc5ebd763f08
41bd946b6287d31fba3d47a77f4a4b5cd441b990
76341 F20101221_AAAVXH miller_m_Page_037.pro
884be7160731369e4a44952baf4eea18
99a629923b9446e812504791ca4d4572cf9f4d10
995 F20101221_AAAWBZ miller_m_Page_085.txt
6b9d08995d19a2a9841f97ff928942e7
3820fc502fca2166474edcc571b4e0e5d22da631
35550 F20101221_AAAVWU miller_m_Page_045.QC.jpg
da526830815b22ffb8135fc99c8727e4
8490928f73948a79643c689b6fcc1afc29874558
F20101221_AAAWDD miller_m_Page_094.jp2
5d31ba414972fbf7542c82c56375dfc1
da32560e819410f2f7faf673871b5170876087cf
107777 F20101221_AAAWCP miller_m_Page_100.jpg
a4133f77c8989167c22276281c16a77e
3e19f563985153a7b4f42c93b1d451d48b5e7e4f
1051929 F20101221_AAAVXI miller_m_Page_003.jp2
8fed4cc67f13e0a3916454bd717f3fb1
bfc1b64ba76b3dc64181f7ffb5e2f082cf8f736c
5706 F20101221_AAAVWV miller_m_Page_107thm.jpg
a77c01a2595dbe725e42ee4991ffa6ca
268bf590947eba328bde0f3c764616d099f65e56
15095 F20101221_AAAWDE miller_m_Page_115.QC.jpg
4cdc6f6a99f1b666962c628d07024b77
2cbe94b93a7f1c16e63aa9a1457a178e934cb4fd
52709 F20101221_AAAVXJ miller_m_Page_061.pro
860fce98912c4c110dc7f1715075a775
310eaed72e5081759e94bc38511ecfb9304bade8
113159 F20101221_AAAVWW miller_m_Page_040.jpg
07cb481b1258c76a1bd901aa0dba0686
321e25d1a0d5553f66542878293e6bf75242babe
2254 F20101221_AAAWDF miller_m_Page_069.txt
879c29de72e3df112ea6c21cfaea2746
1111da73f4a257493ecbfacf3d8706ad249c2ff6
47783 F20101221_AAAWCQ miller_m_Page_011.pro
09314f987b9d391de3f822214722e218
367429271ca7a88e88d899b3df0f726004452e08
1051973 F20101221_AAAVXK miller_m_Page_019.jp2
a51611570d2265fd538021c256beb21e
f69209d69de75b737eaa0f4cbed7a09c3c67b60a
F20101221_AAAVWX miller_m_Page_057.tif
a668207f1bbf8430bccfaffb412a2509
c8a134b27e94fe53d7d506a6d7d5e109325d4cd2
56010 F20101221_AAAWDG miller_m_Page_055.pro
c191743849ad3441d841a1a59d87d60f
6bb726f573754d417363122254295beb2c6d42c0
1051917 F20101221_AAAVYA miller_m_Page_088.jp2
2461eb0d751d515c482f5072d7d48bb2
7175120cb4393e3acfc787ceedc665b8c58fbd6d
110090 F20101221_AAAWCR miller_m_Page_012.jpg
5ac6b61ded248d06512d4af8a38c7855
1c040c93ae6c67e6262df173ce3d295fca81982e
1160 F20101221_AAAVXL miller_m_Page_076.txt
135b7d15056510c96a2d4b465b0a645f
3a6a0afdf1bcc01a55481cfe88e05d026b4942d1
57066 F20101221_AAAVWY miller_m_Page_089.pro
c151d44593ebd3c113d08d7faf465562
42319229e98b1099c29e050bcafe81f1ea9b4c97
35014 F20101221_AAAWDH miller_m_Page_022.QC.jpg
187c5721cfc4a3a615d47923af1f5229
43dcdd65676dc934f8554a9e6d917f5c3eea9a85
16297 F20101221_AAAVYB miller_m_Page_084.pro
b48a9c6c5a8326d9a16771aaccdaa38f
7ceb6561eb0eebc9f1eaacee6e783f5a131af275
F20101221_AAAWCS miller_m_Page_020.tif
9ff920f1fda83df5e79b1c262d4f687d
d727063c844215feb620d38f3f40c1c733cf0d30
13679 F20101221_AAAVXM miller_m_Page_105.jpg
f16dc4b5f6a49e394bbe54c1cd86914f
cfe3d8d0e0a0286cc4ed0af5d3745b9d2b3ff7a4
34167 F20101221_AAAVWZ miller_m_Page_061.QC.jpg
a02381b277c58ee8fb98f7ad2fe30140
d2018e4279cb9887a0f52b001354eb945e5916c4
36021 F20101221_AAAWDI miller_m_Page_003.QC.jpg
d979e34d209a676e934c574eb2bb4174
1280f117cdf347e3f18b7e15f6e686f1c37ee8dd
F20101221_AAAVYC miller_m_Page_102.txt
57a4a2d30a7f5a3623899babc724f5ab
21b246200c08515d10b8eff63e6fc0c9481989f3
936561 F20101221_AAAWCT miller_m_Page_083.jp2
dc1ffb21379a7b38aebcb6b32acb1a37
db389099cfa9acb954300aef49b4bbbff41012c5
54692 F20101221_AAAVXN miller_m_Page_028.pro
efb9c0e4bc4ecc4ce6f39197547ce150
ec817f23f6ff8392bf26b88027dac8f2656f78d8
31458 F20101221_AAAWDJ miller_m_Page_043.QC.jpg
05771a2bc9da7ee68247fb3765aab494
ec465f13464020a988d409177f2432368f6d4df2
2175 F20101221_AAAVYD miller_m_Page_052.txt
d0200beffb9862ee679b928af325af4a
f8686a823362c39134446bb7d68cb964f3ff32c7
138292 F20101221_AAAWCU miller_m_Page_113.jpg
6bad4de9789b42b0c2da4e1ca7d423b3
33b73c1dfaad0c1742962af5a729d2d69e3e5e8b
F20101221_AAAVXO miller_m_Page_056.tif
f5e5feaf3b86152c988927749269277a
3d973527f13e2fbed561ad6dc1df1e7cc0457692
37566 F20101221_AAAWDK miller_m_Page_053.QC.jpg
ee28ff4024ef615132ca041535335d48
93de5bc4c8cc006aeadaee6f001a1c200cc5d7b4
F20101221_AAAVYE miller_m_Page_120.tif
af79e54c9ca780b6e60e98b917eaba67
cbbe5597ca6a722884b2c16c65d617e475d6afa6
36503 F20101221_AAAWCV miller_m_Page_038.QC.jpg
a9c99099b88d2ce3718f25ce9c1479f8
ed071bcb6b00e3a869dbacce7ff3df969eda50d5
F20101221_AAAVXP miller_m_Page_014.tif
9c68efadede86fc74508e6d275c0983e
d91fa73359f82e1834ba7de36b29f1058bd7c8f5
F20101221_AAAWDL miller_m_Page_085.tif
a2174284aa8a9c4c624365bc823b87a9
5d528603c42ca5c259118edbf39f86c2611f5272
1051912 F20101221_AAAVYF miller_m_Page_051.jp2
ac70e34c007932424b893e28ab33515c
40981b5cf0c4e77302f29ab6ad98cb67c75f00c3
F20101221_AAAWCW miller_m_Page_090.tif
6825735b62985c3bd8fbad51b84c56fa
fe5fa4b1bfd61de53af9600ae81b4993110e2ca4
582414 F20101221_AAAVXQ miller_m_Page_008.jp2
f941c43a3e69ebffc168228738b4cff7
f647dc742c571a3c157e7e38e7028bd33c6b5701
574055 F20101221_AAAWDM miller_m_Page_076.jp2
f6a411dd95807ab707781f3e5cb2ad0c
f4bdeebdd84c0d0d505b9e47c96cf11e18d416f1
24977 F20101221_AAAVYG miller_m_Page_005.QC.jpg
79f1dfc63cc55b44f204b4250846c456
4a4c798b9dd5b3732cddcd90c8ad98d6f13f8c4d
9044 F20101221_AAAWCX miller_m_Page_056thm.jpg
5ef4f6e253019b3967da832d43a27b34
642a7cc7a2a14c0b012ea3a3bb81c21cb180b047
124270 F20101221_AAAVXR miller_m_Page_124.jpg
8a0ab87388dd5bbb41ff58eccc024d0a
ea5e9673beeade84c1ffe903fe7fd88d4d99e8c2
113422 F20101221_AAAWEA miller_m_Page_016.jpg
d9f12e5a29ecbc8332a93e259fbdaa2e
33d16ddc6499755f5ccb6ba191268be97a462d2b
1623 F20101221_AAAWDN miller_m_Page_080.txt
557c4288c476829b219aef18f5a6d9fb
364c582453521df40faae5ecd1f21a74fb1e9c6f
F20101221_AAAWCY miller_m_Page_064.tif
42c3a6b5611e3a19c6e46ce72e0c2282
1bb9ecbd18111d78e2ab58f5715879dce6626c71
1051939 F20101221_AAAVXS miller_m_Page_030.jp2
fe23bc4b8872fe297325186729c2a362
da37390ed3f4496d7ba6938f34117776c8233a6b
111392 F20101221_AAAWEB miller_m_Page_017.jpg
f9dce9fe316fc00a67810a92e0f6f971
e13ff8c6c9357982e1598eade93ff8917d2a8243
8669 F20101221_AAAWDO miller_m_Page_101thm.jpg
d32608f62492678475e24609776d7188
e01cf22194a0187ea1ebcaf1a4dfb4737cfc7cf8
35857 F20101221_AAAVYH miller_m_Page_070.QC.jpg
3423bc661d81aee1ee354c206bb9cf14
da23ff520e8307323123f0d9e71dc0d5d2c4c142
F20101221_AAAWCZ miller_m_Page_084.tif
62762d98d5015ffe3f528c1980a999f1
c403cf5cf048a417491b35af8bec7adbd9595248
56538 F20101221_AAAVXT miller_m_Page_021.pro
139bb0f3e704b14f7b8400950c672d9b
034bd8b833cc9530cce6aa38b947e6d4084e0db8
100436 F20101221_AAAWEC miller_m_Page_018.jpg
34ea4b373dff9dead61ba8974a95c05e
48db8935166601ce9e9c321df99a741780bac518
8815 F20101221_AAAWDP miller_m_Page_051thm.jpg
fcc9943623aa7235754b6bc1a57b4560
88fd47e256bb3313f0ed7b0c1f57856acc2c6d28
F20101221_AAAVYI miller_m_Page_046.tif
8386f5e7d6663b9f1e27edb6fd2f188a
1676bcea3ee87b2afadf8c5b34f4fc4be93b7685
32909 F20101221_AAAVXU miller_m_Page_049.QC.jpg
395b7423c99070572664bee72e325b38
dbb60555c08415ced7f6a22debe0463784a69551
112738 F20101221_AAAWED miller_m_Page_019.jpg
838555eaa3a515285883ead3bf0cb068
31d06b32f32659d996172ef58e77806cbb4ba314
146167 F20101221_AAAWDQ UFE0022335_00001.mets FULL
c3c7546a9c4221e7118d40fec022f30c
80d05b55df0bbe97e66dd52e87bb20c57c9b3abe
7899 F20101221_AAAVYJ miller_m_Page_110thm.jpg
c7c8b52211fe0012b322a435c759fcbe
956dc53a2766e49c906260df8c0a764e68053369
2306 F20101221_AAAVXV miller_m_Page_039.txt
16c5f2f71930955bd8c2fbf2d670c49e
6641274178e672c919608a32bc2485a265253e23
104398 F20101221_AAAWEE miller_m_Page_020.jpg
83218eb1fa697d7a44f54f61f56465f2
618f38748c02838b05e568d58bc6380e3da2e76d
F20101221_AAAVYK miller_m_Page_015.jp2
d531937876076d5a61bdd38c83982f47
e8bdebf91fb9832ecd4eb32797253a6e5f69a521
F20101221_AAAVXW miller_m_Page_022.tif
e16de0b6c335a3c597ec3b4739676155
92bb665282ac4b647db8ebcb0b87a0e9435297e4
112317 F20101221_AAAWEF miller_m_Page_021.jpg
cddb92c546b06d7f5ec1f4234a8f24d7
988b8a6ce7c18a96743f9ce2961ba34116704b9d
1348 F20101221_AAAVZA miller_m_Page_105thm.jpg
e693ab5c88c2daaafab18c256c73dd69
de87be49fa3b30075969829b382488131f743c1e
6876 F20101221_AAAVYL miller_m_Page_004.QC.jpg
ef2abaf6d3b8c7788e4b5ef0d6912034
2630614f463d68aca68d05bef4d36c1280068fc1
F20101221_AAAVXX miller_m_Page_066.tif
9ef69a91dce4ae6eb3949126c3daba98
a9dc152bda0d87b29d3c11a75143a4f556688e25
113259 F20101221_AAAWEG miller_m_Page_023.jpg
dc620d44e48d3b2c73d40c97940e7968
174fd7a998fac090becae05971d91dbfa38653b5
441 F20101221_AAAVZB miller_m_Page_001.txt
5a72d0380e7d3af8549dfd78fe659508
fd31279f48a6a32e4d0db8c437b43387283d678f
F20101221_AAAVYM miller_m_Page_049.tif
f54148fa6192437252b7632686c3fa0b
1ba16de24168556ffc880c410dd14089f227698d
2202 F20101221_AAAVXY miller_m_Page_055.txt
6196d3d4a22a3bc2c37c42617707099d
4558edef9e5b0d495f6f10978817e364808ea002
108405 F20101221_AAAWEH miller_m_Page_025.jpg
f770026fc08696935ca43f78e99da2f9
eb42be6b23e20b642719292c72c8e098f9a348ac
1051983 F20101221_AAAVZC miller_m_Page_061.jp2
3bc6ce0dfc12f1b53cf9d5a37d573824
6bdcc3ef6c9b781bccdbdd0417f888e1592ab3e2
24194 F20101221_AAAWDT miller_m_Page_001.jpg
0c94c5263b3d56a4e305854b96e230dc
1c681639490cf632dff52d46a18143363d5ac58b
2245 F20101221_AAAVYN miller_m_Page_016.txt
1a4e6b86c81292e41bc7cfeb940e4667
48636109b76c8b63cca269cfbd27932ecd9002d6
9242 F20101221_AAAVXZ miller_m_Page_073thm.jpg
a1ff4877b835a7ee54acaeb7e2cdcbfb
ef638843be809842e47766cf0e356e493f78d4d7
112979 F20101221_AAAWEI miller_m_Page_027.jpg
7fbfcb7a4a134ecb2084e2450257972a
602f91de304b848793e97529bc47feb59ead37ce
1010 F20101221_AAAVZD miller_m_Page_048.txt
c03addad01cab5171a3e89e5c17608de
eb7a1c1bfc1a6ff3a786f69df1348997c9d76d42
110804 F20101221_AAAWDU miller_m_Page_003.jpg
8f6bc8b7db4fab97af299a55caccbf1a
a4300b005b5f1e3be805212f9232f8491f61c920
F20101221_AAAVYO miller_m_Page_025thm.jpg
7d5b3eb2fbf9fc05633369471ab2dee5
daf12f2a98bb3af9984c1fbb84db050921ea1071
109367 F20101221_AAAWEJ miller_m_Page_028.jpg
88a04dded378b59edf6545b3324bdf64
def7eb9ee1a2ddda0752d1f02f4ca596ca97e6f8
2122 F20101221_AAAVZE miller_m_Page_029.txt
10a4252a226eb752e4befaeb567bdc0b
a53fcb8ccea84126092845f92bae1c177e9efb0a
110178 F20101221_AAAWDV miller_m_Page_006.jpg
cddbd2ce3a41a88edf92923e3871cf30
1e55937a1a121755f11444d74da6117b9dd876a6
F20101221_AAAVYP miller_m_Page_002.tif
15531874c4e2091a82682e7d17b6d757
9991dbcd8b1a77073c95c1c78b22ae988691c745
114882 F20101221_AAAWEK miller_m_Page_031.jpg
57f4d5723936c32d718f36e320fb1010
f29f9bed40e6345d4096c6fbca7373f56ece3883
21292 F20101221_AAAVZF miller_m_Page_004.jpg
56831d5ffd84148c5f19668a56cc81ee
63d6d844db2eb450e6c0393cfecef842b74a0a74
26408 F20101221_AAAWDW miller_m_Page_010.jpg
a3f4edf9cc4102bb595dfc788ed71c94
d297a81ee1de40d730202c94bb36b0208f4baea5
F20101221_AAAVYQ miller_m_Page_021.tif
35a1de4317e82cb5cbec4ec5db3d3dcd
0f24325a71748e301c4ecd95d7e11a1c5ee3b10e
108109 F20101221_AAAWEL miller_m_Page_032.jpg
a61026cb8d7a3fe1247699e47534fba0
a0f69e9c9b180d499b345719eb571f75232fdd0a
54100 F20101221_AAAVZG miller_m_Page_120.jpg
ff23db29a331cb0f65bb0a8c3fe0e360
0a8a3e93be8a82cd5d913c4471a10f81b0c425e8
97141 F20101221_AAAWDX miller_m_Page_011.jpg
3285f1a9611e8944dfbd1b787d843174
b4b8d6cde66059f75a5dc14bbe0ce4161922c775
2292 F20101221_AAAVYR miller_m_Page_036.txt
5835718fd78a05a5c750cb144ee87968
7f283744557dcd23be6e9e8fb6a6206d6414bdd2
F20101221_AAAWFA miller_m_Page_069.jpg
7a6755b1aee01ecad3018a9145b67b33
1d258c5eb1fbd69266f409d67eb182d62b7af4c3
116930 F20101221_AAAWEM miller_m_Page_039.jpg
907266df8ffc47e902a40f967493bf97
917691d2fb298a655b72ffa33a14d775d58f02ca
15645 F20101221_AAAVZH miller_m_Page_121.QC.jpg
c2eda232f876ad04376bb27b1ba27d9a
af40cd90e1637e4e86b919cf7d9381f9f52bac11
116685 F20101221_AAAWDY miller_m_Page_013.jpg
daf8f13cf6772364d82851533eb442b5
14793442089bbdfc79ae1310328cc94f81e119b5
3404 F20101221_AAAVYS miller_m_Page_005.txt
5f9a789bc69232c6513da7b98e80aa17
d3c14f89106eee0985ba8c29d74d19c3eef5285d
107377 F20101221_AAAWFB miller_m_Page_070.jpg
22600d7567850e695761a007f3b992de
95b06008893cc96b2a1d26e903e9fa12cab86977
39516 F20101221_AAAWEN miller_m_Page_041.jpg
8a6e3e61d7a2c91decce7c9c92197509
2f7e09b60de4a2cf1ba7426d3a226e9262adf281
110170 F20101221_AAAWDZ miller_m_Page_014.jpg
2d0cc740e2b99541013b0f479864a242
554014e18b7ec919610dd0b829b1b029401fa0ad
1051981 F20101221_AAAVYT miller_m_Page_060.jp2
478f654ac0413826a25a162b3497d2d1
30b646c28347fad96065e7fce5668a55e9b4e522
113673 F20101221_AAAWFC miller_m_Page_071.jpg
c9881260648606a787b8f6a94c03e8c8
a10ad48292cd68e3a2515f8aa2a7c881d63a18db
99551 F20101221_AAAWEO miller_m_Page_043.jpg
f57014aeaa7547d82dc5e22ae5c177c8
bad9e8242b2109645bbf99c7f75202172e49e11a
544495 F20101221_AAAVZI miller_m_Page_077.jp2
cd2b733534f4cc0df1cefae4dd889fe2
56e67f2b19278d92879612f8e7fbd993fe0b838e
2147 F20101221_AAAVYU miller_m_Page_117.txt
903015d4d155fe2aa7f7533d5e255b01
3a043a7d4923bbc259e6b5dfc32fac0c1b11abb4
113749 F20101221_AAAWFD miller_m_Page_072.jpg
25a569b4dba801cae8954def525483e0
9a8b6c50b528f6b123b0285a4990c1d2f2c23682
112963 F20101221_AAAWEP miller_m_Page_044.jpg
1b00c454eaabec84f2585d73692343b0
f836c93287c27cc6476bccba7224155dbb51e4dd
F20101221_AAAVZJ miller_m_Page_099.tif
9aa2e75b31fc9862ad014ec10ca2c53b
f5bfd60e7b743636056197810f34741f217cf080
111101 F20101221_AAAVYV miller_m_Page_030.jpg
15f604b2e40f628fb2a19f2ea309e602
2af778c9e453721d787bc537a1e3bac8307797ef
63902 F20101221_AAAWFE miller_m_Page_074.jpg
1e9ded63d3cb79da209fe84ae224b019
e6b28e6d300b62f6681bee782edb643b0cffebdb
109505 F20101221_AAAWEQ miller_m_Page_047.jpg
f80b29c846acc7501e2b442ce6ace5fc
65f6b680993778fceedac136c361da211ab27202
8904 F20101221_AAAVZK miller_m_Page_102thm.jpg
7fcee6ea6d83a901f28d6c3be15e6639
4358c7abb003733c802134e322f2adc777d6e993
2802 F20101221_AAAVYW miller_m_Page_123.txt
5ee515b0193efb84b9a133fbf1119ba8
44c1890c71bcf074c8a47c890be114178e35ba87
37256 F20101221_AAAWFF miller_m_Page_075.jpg
db8a298a0e9e923a9e696b10f1165522
003ff1526b222204c4bc2b5a9c975e7b7fdbd343
54069 F20101221_AAAWER miller_m_Page_048.jpg
80b9469e927d6475596cf07f40a9d564
a4e45a6c771803e35794ad9e5f9e6cca8d7bd883
33660 F20101221_AAAVZL miller_m_Page_087.QC.jpg
d239a2679ac0f4b8236dde4e4cf62b45
98055b7d4362adeb039bdab78425f0364be31ed5
F20101221_AAAVYX miller_m_Page_079.tif
7c199e27d8d92d55224bb5ea3c3bffe0
3cd89def8800ba9295bc8acd4bc36ecc0d64ca3b
53793 F20101221_AAAWFG miller_m_Page_076.jpg
24e18318603fedcecc2dd01a566bf4e7
5937ab715f5ef598e851ad80a6b7c088d9ba453a
54985 F20101221_AAAVZM miller_m_Page_067.pro
5222bc74ddd29c1ccf73f78995804e82
b0994e6991ef51d4875a199ffa66eba6e15ff120
921 F20101221_AAAVYY miller_m_Page_125.txt
20111def8cc9e13231e184767e48025b
21ca60bfef9a98bfa957b34b2e0a4e35852ca760
33051 F20101221_AAAWFH miller_m_Page_081.jpg
7328e56e757ceb3b740ae3aba093465f
898af975f67777b3193a6fb27e16eafbe2e081c9
112814 F20101221_AAAWES miller_m_Page_050.jpg
96f6bc4724b5877f1182a01311574d76
a2e6f5b7925b276317cea04f4c8a35ac73024300
F20101221_AAAVZN miller_m_Page_125.tif
e357871e9e18a9d89777bc43240fe1e8
8e06cf260a5875bcf8e631abd77c6b1fcac251be
7763 F20101221_AAAVYZ miller_m_Page_011thm.jpg
ab3efa0a91e5fc7ac361c33a15ede315
76bd8ba919504296e4cdbadd4ef1f27b649b5928
40242 F20101221_AAAWFI miller_m_Page_082.jpg
3b997e5f637d1c5390a49970033b71d3
810652226dbfa58976fec559e494da374b25ba53
113184 F20101221_AAAWET miller_m_Page_051.jpg
865bf540e0ddb7acd02d9635cccf8e7d
d3bab363c03f1e8958df1d6a8cd01760fc9f9da3
1051967 F20101221_AAAVZO miller_m_Page_101.jp2
49ea36886cc97c89e5b9aaa4b4255bb0
42347e4015837721542cc20b1dd0080d11acfc26
86289 F20101221_AAAWFJ miller_m_Page_083.jpg
cda5d3d125ce4fb1c9b8d093d1132bee
0e49119e1e828a1cb9013f7c4128f9918b407056
115756 F20101221_AAAWEU miller_m_Page_056.jpg
16178b6ae6c3808556ea3af1dd8354d2
a5f7c6676beb8d59bf2a1cafc61ceb585c9fb50b
36293 F20101221_AAAVZP miller_m_Page_023.QC.jpg
3f7c3febeb032c1ea49fd4b9f426ab55
864857d8a2a588383ccddf91ad0c334922aac8ea
26560 F20101221_AAAWFK miller_m_Page_085.jpg
1a1c56e984f607e51b1df1b0b795b8bd
108c4965cc45e351e9388f3413c8ae0292c503ab
105011 F20101221_AAAWEV miller_m_Page_057.jpg
e01fd13c0cf2b7b9c6c53323522d0ed8
849630006cc28478f19feb5405d1694194942144
34752 F20101221_AAAVZQ miller_m_Page_032.QC.jpg
f461fdb6662819d4d0e7acc69fdbedd3
3bb52e35617e80e7101253c66fd54dfd0c71295d
27747 F20101221_AAAWFL miller_m_Page_086.jpg
36579906a0b476bfee6f268a6846a918
b2d46e572013f9f8ce63445e13d5893fd1c3d38a
109683 F20101221_AAAWEW miller_m_Page_060.jpg
d9c27d6ea6cc67dbfea1b8536b7702c5
88ce495e39eb1428d8cb5f22580df8fb1c00eab3
F20101221_AAAVZR miller_m_Page_056.jp2
eafbadd761ba2eb79915acaed9c0acec
3998006b33c2fc93dd3d4e0795348113c49cfc41
82790 F20101221_AAAWGA miller_m_Page_112.jpg
2fbee8fac0021c424527810e106e32ed
fde27fb700a9aefe0a8eb406096643f61f29242e
106033 F20101221_AAAWFM miller_m_Page_087.jpg
5aa717288068b68e030c26e281d1618b
fb107e84a0e61e8cb9869082afd83a0c5d437490
107683 F20101221_AAAWEX miller_m_Page_061.jpg
0edcc109980064523df512d05ac08e92
51f6258f510e1a849fc644d89f4c6a0d7443e561
78721 F20101221_AAAWGB miller_m_Page_118.jpg
248cca508b92e04b4b30413c6e275e60
003736b5995492eeee180492424f8fbdfdaa378b
115811 F20101221_AAAWFN miller_m_Page_089.jpg
584922f7ec91d28832c810c840b57848
8282a7605e111dad3377dd89228e4bd270949d50
110752 F20101221_AAAWEY miller_m_Page_065.jpg
ba7b96fca23a05a9512d8d95a74cdfd8
3e233135f1d6444ff9cecf503fb96401a1e156c2
36487 F20101221_AAAVZS miller_m_Page_067.QC.jpg
806a11c68944d8a478b9e7f32d69ed60
13007c11f5e84319666de3905af42047d78f3eea
54706 F20101221_AAAWGC miller_m_Page_121.jpg
28ada2e1e3692aa79b5acc3458132750
f5a385a3b7ffd31d8ada908320806fcf4840c9d9
107839 F20101221_AAAWFO miller_m_Page_092.jpg
bcd20851b046278ed2269ef4519312a6
e43fad3cee66392c956d5484eeef4713b29dc798
110277 F20101221_AAAWEZ miller_m_Page_066.jpg
66af979156a50744bb78c33c9ec47efe
4ebe05233680ef967811c309e19be488dad5b163
475 F20101221_AAAVZT miller_m_Page_010.txt
d490b03da8a82ba62f41c2b5c8af9860
96c82de54c0e1ce30c07a65b0e647e243da586cb
142818 F20101221_AAAWGD miller_m_Page_123.jpg
863bd060ad6d3d6e778cb142c67a19e4
7037b5186c1b2b331b739e3829fa73c51e372c41
113795 F20101221_AAAWFP miller_m_Page_093.jpg
da78ccab7cba9d8c990d8706ad8de608
ab1fde9babe779583ab41f15f85c4773d8df0e03
25313 F20101221_AAAVZU miller_m_Page_048.pro
0693126b9e1448f000d7bb8a5f540a23
ac7c645085a3259bba22a60e329fa09687728484
47025 F20101221_AAAWGE miller_m_Page_125.jpg
6d43fa80998cc0195fa5c938c22007cb
22113a41ffe4d7b3ddf3e7284b792b2afb356f70
105706 F20101221_AAAWFQ miller_m_Page_095.jpg
4b7f6370498a0a7d808c0eba59b58653
fa69a28a95b90e75be7dec277786a7be8bda8e35
2455 F20101221_AAAVZV miller_m_Page_073.txt
cd4b349ed1509d67ef513af517eb55fa
0bb61d18e4a04a90017cd2b2bf32c2259d8153ff
24965 F20101221_AAAWGF miller_m_Page_002.jp2
d689ff8d5868cedb7e1ad292e1a41c59
7d9da8c8d10c6fa853846b7b88f09671c2d623e6
115922 F20101221_AAAWFR miller_m_Page_096.jpg
53546235e4b2d15a9e77f47dfe99133d
11c2b434476b189f0effbb1269ef771aff701b65
F20101221_AAAVZW miller_m_Page_029.tif
d3fa14c7b61def7c8099f9b525d59389
acda428cb1f8ca3b606ea7f7a1be4b8a9c42b06a
209791 F20101221_AAAWGG miller_m_Page_004.jp2
b53dcacb9b76d238aa5cb97f67f8a360
60cf02b3843c9ddd9642cc9028a0005437d4d774
109893 F20101221_AAAWFS miller_m_Page_097.jpg
e3aef625c4ea76c9e8454777e08eff52
2ca193739defdcbb1e2e32c861de7440822fb572
2008 F20101221_AAAVZX miller_m_Page_057.txt
9e8f080e85c7dc06cd6f92da238cfab9
294e55fd327dc0118ad94a03fdcc229962c424c9
1051940 F20101221_AAAWGH miller_m_Page_005.jp2
392c496b70344d9a1fc48b4e349b3b48
9fa67fe5e6bf501ff7793cb30e405ec5fe0f4e8b
127922 F20101221_AAAVZY miller_m_Page_122.jpg
2dadf80950a574a93b4b8e79885d2f82
2a1594207d58e2e5ddd8ff7b68e0ce0c2739a457
1051976 F20101221_AAAWGI miller_m_Page_006.jp2
5aa933d2e4e867399d7a24aeba9e8f08
615d583486325130d1a4eaff7ede830347f5a3bc
111778 F20101221_AAAWFT miller_m_Page_103.jpg
4b63b57b35171363d311af1697319744
58c6160e7a0940ce3add4edfa3c5bf202254b0d0
8719 F20101221_AAAVZZ miller_m_Page_061thm.jpg
b4711f0cbe21b22cac119d9d7f405fa5
4298d11cb682d42e0a54844eafbc1f2314249c62
1051959 F20101221_AAAWGJ miller_m_Page_009.jp2
d6c25e49495683ab5d051426ebeeb941
3a8f9d7f39a92d9cee2009fd7356df12165699c0
110138 F20101221_AAAWFU miller_m_Page_104.jpg
316a2942711851a63b9bf115343c7c0b
16735f876f8aa203dd24e7282f0324d61a317b44
F20101221_AAAWGK miller_m_Page_011.jp2
cdd48b2f33e47185c81c0cbf31eedfa3
ce36e370274f34dd6a6f3b0188a966f4fab58633
129536 F20101221_AAAWFV miller_m_Page_106.jpg
f0711fda072485d22db73763631ef914
85d5e622375129653ab479bf939f4333b2e9e61c
1051918 F20101221_AAAWGL miller_m_Page_012.jp2
b152d9d6dc416299f379c6120f5c1911
be77b6418a7ea827098bd4560b1ac2c7def6840f
97835 F20101221_AAAWFW miller_m_Page_107.jpg
404553640dd48df1e3d8b7a433bd05f3
3364bc59bb889305dbc137cf29a0dd395d0c23b9
1051975 F20101221_AAAWGM miller_m_Page_013.jp2
9e44c6fd171890417940b06fca02da7f
5cb728e36f370926fc2c1f8f999cae3df78c765a
65068 F20101221_AAAWFX miller_m_Page_108.jpg
b16c66a959018d2811965db34c27853e
19351e7af3ce7ca0c3c5463af4bcff9c7ac1ca76
F20101221_AAAWHA miller_m_Page_043.jp2
dce6aee658495e801e708b8255fb34f9
ce0816bb5662d43695737d84d5e4d3895b5cdeaa
1051900 F20101221_AAAWGN miller_m_Page_014.jp2
d8718e9f5533f8579129942bbf211344
95207637fc4aa8127beda06c6454783412551362
72815 F20101221_AAAWFY miller_m_Page_109.jpg
4f136b3b2199669f2f763c95d97c2968
13a6c118590427e9c19535ece44c85c053b6557f
1051968 F20101221_AAAWHB miller_m_Page_044.jp2
55bbc9d73afc817a1f4b94b1cc2021c2
ebe32ec00593151fcaac72bf2448b7badaf7a8cc
1051974 F20101221_AAAWGO miller_m_Page_017.jp2
d7835fa947a5fcc392f78fc080a0f807
64a1b89c3861aaed37d514244c68b35347b188fc
35788 F20101221_AAAWFZ miller_m_Page_111.jpg
26a1376d3340bfa9e36c68ec5dae4890
299da4dc77440ce269f8aa4842f90d606029a55e
1051964 F20101221_AAAWHC miller_m_Page_045.jp2
e56c0311785c0fc87087786123ec81c3
6ee05731eb93d14aee583a221d01009e1c1c51c6
1051930 F20101221_AAAWGP miller_m_Page_021.jp2
2439b39b6c19936f1860a9400700bc8e
b0dbf3c82c7331578971f22046a6cf650c749616
1051936 F20101221_AAAWHD miller_m_Page_046.jp2
cfd9973e48d7fa14c4837e6edd99342f
62977721421162af49ea463a94737a4a76632eea
574886 F20101221_AAAWHE miller_m_Page_048.jp2
cfab58d28e88772ec4256364d2ffbcf1
ebb2bd16e386a1b1fd43731fa2024391cfb6f9db
F20101221_AAAWGQ miller_m_Page_022.jp2
46082fcc2cbdc9d6742af7e1bcecfbd9
ac82fa3975602b26967bdcacdee4122f23f16999
1051985 F20101221_AAAWHF miller_m_Page_049.jp2
9d9eb722db5192ea435c4a59e2855e74
c9931166bbdd5cdbbee1bb5b9c287f4d23808629
F20101221_AAAWGR miller_m_Page_023.jp2
36661cf175031df6ebf1c7f1429ae9a4
e99196a36efd6d06dc71a9c4f73653a3f7b1ed88
F20101221_AAAWHG miller_m_Page_050.jp2
9e976588a7a31cf8d1b52d919fd0927f
c5aa27a7e44a3c0b3a080b646261f31da3290c89
1051966 F20101221_AAAWGS miller_m_Page_026.jp2
fd29cba0c728c03df5c8ad6ff5917785
5da5dd08d30ca00a6ab2643f570e3f72e88e4e82
F20101221_AAAWHH miller_m_Page_052.jp2
0f2785d498bf3cc5730795e5627b2c59
ba0469226afc38c4038eeebfdaac0c40341d4621
1051911 F20101221_AAAWGT miller_m_Page_027.jp2
d4aa06f1a7592c22e1a7f45a90684dee
89fb4e61e964af6badc3fb57bebfbaa154f683fe
1051945 F20101221_AAAWHI miller_m_Page_054.jp2
f9d602004f61e7bfd3b1031b9a95b1ec
eee4f8b018b5cfce14c93778ad0f551ef184f248
1051978 F20101221_AAAWHJ miller_m_Page_055.jp2
a9cefee91d0ab490789be3d25999d0ad
b8f1359f4f6160d02b8633aa9775fc88cbcc21bd
F20101221_AAAWGU miller_m_Page_028.jp2
776918e316181c2291f9664a9a5db23c
0a6303a8f856919a41633c57e5a1083ab55d90e2
F20101221_AAAWHK miller_m_Page_063.jp2
4727629c97590a74efe4f2491255eef9
27327bfa36800f5103cc7ba7b3e7aef72acc6b06
F20101221_AAAWGV miller_m_Page_034.jp2
9efa9549809226b4922f2ef80c3ddf12
786070007e2a5ed191bac345c1630441894c872d
1051955 F20101221_AAAWHL miller_m_Page_064.jp2
a17553e381d1d654cc99ca0e5a1376b4
7ee22ec6352abcdc27b2a353498280c880f68122
F20101221_AAAWGW miller_m_Page_037.jp2
b769b2bbedec4cfc621177602fe4b075
7756536210ce5517db5b3bce1f91d5838878e948
F20101221_AAAWIA miller_m_Page_096.jp2
30cd22fa8d1c532898cd4c98ee27c1c2
481d00727bfabd733cc291ab63aaf7a46befc895
1051962 F20101221_AAAWHM miller_m_Page_066.jp2
4ad29c5115cf1386e6a0c3841136ee31
82fdf5a9fb11ac928ce4d0873fc15ccf8f18164f
1051969 F20101221_AAAWGX miller_m_Page_040.jp2
0922811145a52bc6b552ee9c1acdd488
01d11f8efc16aa9226538a381c7952f45543dcb0
F20101221_AAAWIB miller_m_Page_098.jp2
5deb17cd759a9270a3f2aa49d12d1158
b1b73af81a4ecd2a17e8490347e7e9b377c9ce06
F20101221_AAAWHN miller_m_Page_068.jp2
c2d87fcfdd7ba61221defd92784d60e2
11e6498d8bfce69604d29292ae4bd3d6a7d68eef
422425 F20101221_AAAWGY miller_m_Page_041.jp2
d10cae7edcb34fe8b7e3f855c9e7f3ef
3240140998ec93c5667f730df56cd554c398fef0
1051961 F20101221_AAAWIC miller_m_Page_099.jp2
e0d61dd7a3713866c657016fe0c547d8
f013efd61671b7fea4c5b18a11ab8218908f937c
F20101221_AAAWHO miller_m_Page_069.jp2
be364dbed7bebe52e2b23d40fef66d94
ef1e274b220106ea567c8238a3b87bbf1b582737
1051977 F20101221_AAAWGZ miller_m_Page_042.jp2
ff4095a870e1d14858f98576bdb1fb77
ef918c88adfdebce0ebe25852405dde8e7da6ebc
F20101221_AAAWID miller_m_Page_100.jp2
63d5e4509d4951b784f6743af2e08ded
0604b42e0a17762704e881057f1ab49bf471d661
1051931 F20101221_AAAWHP miller_m_Page_070.jp2
942690b924af1cbc9239b602f53f154e
65ed7539212d83afc1814ffff3181b8ac4618be8
1051894 F20101221_AAAWIE miller_m_Page_102.jp2
920c66378645105587949b14bb7a1449
8c42d30ee1b1c19ead5ea6dc2b026c84601b0d38
F20101221_AAAWHQ miller_m_Page_071.jp2
c32ef53aa523c4b5775b348f80854ba3
111b866ec9fe571fbdc249ed4c8e288e73040855
1051938 F20101221_AAAWIF miller_m_Page_103.jp2
5aac8cf0bcca6ea188da6b7fd0b80c72
7c011592d3f689d87721b941d1c470f82a3c0329
F20101221_AAAWHR miller_m_Page_073.jp2
72fc1d582410a34a135ef07600af55b9
dd459539169a289d845cece9ee77f3be7752b51e
F20101221_AAAWIG miller_m_Page_106.jp2
fcecd903b99a90ffc3247b4c6b0b203f
eb66455aa441dd778221234deab30d8195c249a2
692206 F20101221_AAAWHS miller_m_Page_074.jp2
fa7acab63af5357c2601413ed46ca6c0
551bfd8e47612d0d41b849f86d077cab32630be3
F20101221_AAAWIH miller_m_Page_107.jp2
f5e7c68547de9e747e343b7893e008f4
fbbf5b1d9ac86c0451e6b72243254565992c453a
386048 F20101221_AAAWHT miller_m_Page_075.jp2
3adb68ac8110c3f40f906d60d9c2fcd1
78fd8e0fa091e8f258be75c771ef094f94da2637
692552 F20101221_AAAWII miller_m_Page_108.jp2
e63fadee2dfded8a41ec9418515b9ac6
68380a7bd24dde2fd8b99b38fe5baa5d48594853
1020719 F20101221_AAAWHU miller_m_Page_079.jp2
543a3223bf74ab478d3a00aeca272ee9
e66a7bfc1b4cb1a14c3994ddfbb4f6a3a41a1c5d
774129 F20101221_AAAWIJ miller_m_Page_109.jp2
e62de1101178ff16861fc8668538c828
8531068830d9efc9399a9bb37275e56ee85dab9a
F20101221_AAAWIK miller_m_Page_110.jp2
2494234b37c0b48fa49d7401b4e35835
b721f5a9ed99e73dfb76b7ebe4bc3a7b5b6cece7
330727 F20101221_AAAWHV miller_m_Page_081.jp2
48755909dfd197c1e0eec1d858b38f85
d6ed5b54650acd331a078dc39114f5415c4c94bb
365371 F20101221_AAAWIL miller_m_Page_111.jp2
938c36d2c4d5b7d89ecb8a9b46f9e732
bc13a289c73d09761df4d4a690b77a08b0777108
374959 F20101221_AAAWHW miller_m_Page_085.jp2
7a571ecec4b1a2823a9f5b916913fd2c
a86e82bbea9bf37c274790533380ac0f883f809e
F20101221_AAAWIM miller_m_Page_113.jp2
dcb3d02015e0eab8bc6af871ac3d5700
866b6a664f528cf5deac6c5c3bb07e4d8f665383
F20101221_AAAWHX miller_m_Page_087.jp2
ae385ed89b87d374d3836e2427f3fbbf
89cb5682feecf182850c93e351b76be2ccbc4fba
F20101221_AAAWJA miller_m_Page_017.tif
f9e0667573551979cd103ccc279cd37c
727a81aacbf96396e91ca9af33cdfc52c7698c81
F20101221_AAAWIN miller_m_Page_114.jp2
04b0452a330d6ada6df7fad013020df8
8319b03862754748f301ac8c9fd6abca1609f8cd
F20101221_AAAWHY miller_m_Page_091.jp2
c84bfbe6236a92bcc891d307fc7d5d98
5f3979e1617830462cdd7626d9d0015812d4a73d
F20101221_AAAWJB miller_m_Page_018.tif
dfa75f35fe05a9922fb048db97e2ec1a
3215e61e522b64c0f76ba352f9a009db6f3b140f
503276 F20101221_AAAWIO miller_m_Page_115.jp2
71249fea2a016046d00b773ae6ab975c
f0cc642e3af1e3637b9966bfc128096bfd980c01
F20101221_AAAWHZ miller_m_Page_093.jp2
2d38f306aebeb1b00423e9764f79bf49
55468e848fc65fd6fd47dbbd896d0db841de3ea4
F20101221_AAAWJC miller_m_Page_025.tif
04744b326444fe6539b1144bcac00fae
acb50e8aa51fe8dd6512cc49289319c65e1ca012
910535 F20101221_AAAWIP miller_m_Page_116.jp2
655145e60956d70cf9b3c722384ea349
0982bee978a11856797391d1178e27f6aaccb4d9
F20101221_AAAWJD miller_m_Page_026.tif
f37013b18bfc5a57936c468da7d3d091
6b06ee1ae9d56223320dd3ef8f7c6d2ec5a52577
836229 F20101221_AAAWIQ miller_m_Page_118.jp2
c74eb06c5dbc631b6054f05ca2b8f1b3
6349748aa5ee15e6ef0c434c1318454102afc95e
F20101221_AAAWJE miller_m_Page_027.tif
d2a3fb9f8a3ee23fa9bea040177b7717
ba44706737968e84fdcfe5797cacf87bb3b7242c
1051963 F20101221_AAAWIR miller_m_Page_122.jp2
6014b8a415a05be91ce097541557bceb
f2324cb0c2506c35412b7f802cf3e15cd1a0f249
F20101221_AAAWJF miller_m_Page_030.tif
d6f26a39d1f398b78da6534cb0274549
c69396d778017206a4f4fdb35968d72aed87a230
F20101221_AAAWIS miller_m_Page_123.jp2
694c3f816229056e15cd686cf10d0709
e3f90cdf2d540ebc9ebce7923c1fb6f485b009ad
F20101221_AAAWJG miller_m_Page_031.tif
10c8c5840a6e31f87590b2fd80e814ac
f41c72713718dd2eedbfc46877732585408951b5
F20101221_AAAWIT miller_m_Page_124.jp2
d0513bf90e2323af1da2662679993806
0c8199f840ed1e279d0efe72d9334417d7209dd3
F20101221_AAAWJH miller_m_Page_032.tif
394aaf002b48531adf3dc2d7fc59927b
9c4d6f7f12fc2f81e5eeb9f9623bddf3749b5834
F20101221_AAAWIU miller_m_Page_003.tif
466c719091317783b0deaa1313d71d38
deeb4d8be39ec2db2e31ecc0ba2020dda6e1abaf
F20101221_AAAWJI miller_m_Page_033.tif
7f0186b5cf6b4f8fe6d71debabc27c64
e816a4f05d9513c0b57666bc1741b5dc8b4af769
F20101221_AAAWIV miller_m_Page_005.tif
0d44afde2b0354f5f4d4e0fd74f7e829
6e06ebed9ab2972d978258c852f7f8b85a57c22a
F20101221_AAAWJJ miller_m_Page_034.tif
1f4a266472963ff7cc60ce349a6268a4
8f1297765ed6f513def5827057b946ea47f64764
F20101221_AAAWJK miller_m_Page_035.tif
3241a674b7270af5b3406f10c03c3c56
80b7a0ced31f85cb0414f863faf92c03e2eaf6b7
F20101221_AAAWIW miller_m_Page_008.tif
ea0ccec6d302560760232c97686d92de
a060ccf9f90cf4de70c944ec9478c194c22654a9
F20101221_AAAWJL miller_m_Page_038.tif
f3f9702f709665f0c2cfe75d0efc70da
36786f05b748df91e99bc865bd27da613fb99fe9
F20101221_AAAWIX miller_m_Page_009.tif
c368567de29bcffa0ddba7ee5f3b4662
6d97b72a691e43203f51f59ca24a49827b73526a
F20101221_AAAWKA miller_m_Page_069.tif
7dd85e0e3819fd39f14226b8a4d83266
68f343943dd63752d2866ff5e92d31cef81fec70
F20101221_AAAWJM miller_m_Page_039.tif
1b224314c058563fea4da73601dede80
276721bff17761d3fd428ed98dfd69c582353871
F20101221_AAAWIY miller_m_Page_012.tif
390c6b2216ed87ea45860a97a5dfd848
58cf2d7690a7e7acf5959ff28c70cfc8f3c57bd1
F20101221_AAAWKB miller_m_Page_071.tif
9737fbab632e38d0451e5284bb3bae52
b40385d60fe921326165b0bf8acca3a9a33b476b
F20101221_AAAWJN miller_m_Page_044.tif
f851ac4afda5282ea7752cc9c76123df
bd39729e625a5efa890b967d6efa452b7759fb63
F20101221_AAAWIZ miller_m_Page_015.tif
a0753d0e7c726d528563327e4848a6cf
d6fef284bc39f762bdf200e7452fe15d50439686
F20101221_AAAWKC miller_m_Page_072.tif
a3037b89dec6d6d7c950c5fcc79fd164
3e786dee377cdbe41912c04ecec3eee8552901b4
F20101221_AAAWJO miller_m_Page_047.tif
3b2dd7f037f3920475eba80421e7a151
04fee745a4205145d909024629f9495f85116ac7
F20101221_AAAWKD miller_m_Page_073.tif
5e2e91b3f5dc15a252753b0dbb6efe47
703c68c6490bb3f88b7d487d90c63a4057e71e08
F20101221_AAAWJP miller_m_Page_050.tif
3f0ffccdaeb70e1b184e47553b8dd3a7
7775d27f2c2faa3aafa7a7503e229865c228766d
F20101221_AAAWKE miller_m_Page_074.tif
81f94233925bb9a994004adeb542478a
e902989fdabcf5df95b72bcf0f58a86a80204db3
F20101221_AAAWJQ miller_m_Page_051.tif
dfde5b42b9ae6f8bc827ccecaf5c1137
3f4c40b08a1799454880b0bf458636ee4c89dfd7
F20101221_AAAWKF miller_m_Page_078.tif
8b31cbec3be1472afdf4d3d0e6d7293b
f082f82e8297e334b9d0b1d12c24c21187f2dca3
F20101221_AAAWJR miller_m_Page_053.tif
ac01ce2d6971c9a8dd6168237df520d4
4415d59dd97c5e0e1a0afba0ca2c2c95740357b0
F20101221_AAAWKG miller_m_Page_081.tif
eac19431068c5ba1e4c8f22aac5607d8
847c95d04d62c1ff336c76e77df53221a08468dd
F20101221_AAAWJS miller_m_Page_054.tif
600967e7681cc29a9c0d26c8787aa142
8b75ac1df83f3ca66774789affd6cb98f5d159e5
F20101221_AAAWKH miller_m_Page_082.tif
7799ba7643c194277abfa10491f83303
405a558a000b06911acfd330ae1a761ca3964e5a
F20101221_AAAWJT miller_m_Page_055.tif
81bdd62064b6f74d0aee9ba7664f14e4
9471e16b985180ec518702ebb420f6c54196f178
F20101221_AAAWKI miller_m_Page_083.tif
b74967c71c125389d0167cabb250cef4
272f066ee0ec416d270d336e388029fd6b791a37
F20101221_AAAWJU miller_m_Page_058.tif
165334e89d08fcb374a66063d6dd2e4c
4df60c4b0fb5f57d28d2d756fc6306346529b7ee
F20101221_AAAWKJ miller_m_Page_087.tif
ebd3d0f7b6822fe4d78dc1bf86a0ae89
49439aa152f677f420d4d926c6a5135eb01da900
F20101221_AAAWJV miller_m_Page_059.tif
bd741286e407de0d6f0eb8c5c5abdb0d
c3465714e8f7708997b821d4091aaa4f54d59a2d
F20101221_AAAWKK miller_m_Page_088.tif
bd89283076e6b8efac491bcb352c1270
7e0e6e83faee92fe434dfa6e0751ff8e7fe2adab
F20101221_AAAWJW miller_m_Page_060.tif
0a82b05adb92764ee7bb1bf41ac74feb
2df116ec4daf649e4c18fc189ba9b10f7d70b0ab
F20101221_AAAWKL miller_m_Page_089.tif
bcbf4b61ba2e4830efc92d939fd3e553
506d54bb4cb88f0ceb8cc2a73c0e603cdb62fdc6
F20101221_AAAWLA miller_m_Page_115.tif
f6f89486366b2d45abe532e55fd73c2c
b1059def21d127182a1a40449f265d56681297c5
F20101221_AAAWKM miller_m_Page_091.tif
a1c6eb36ad2b8b68cc72aeb3c1cb15a1
885d9e0db552eea7853b1c9bf10678fdc4f8aa5f
F20101221_AAAWJX miller_m_Page_061.tif
dfcd6f02e5b37ce000732f589f6cc267
013aa951347649035a8b8178bf051fc8517229e9
F20101221_AAAWLB miller_m_Page_119.tif
20e8920734cf1b721da0551e0c76b928
2fdbe080fcf14f481b5e857d1d3b6e9673477d7b
F20101221_AAAWKN miller_m_Page_093.tif
151ba57e896899665dc1605ce9d95232
74d4438f8bea9d8cfe6f1fce838f1def2430c48f
F20101221_AAAWJY miller_m_Page_062.tif
94efde3c1665e21b5c9c23fd99b2d302
7d3e7efe521a150a88aa78ebf1b45b2fbad2eca6
F20101221_AAAWLC miller_m_Page_122.tif
e5e08d18a774da6fbb135e515363de22
a6c6903e54f860dfbdb0b52eacee21433c16fa26
F20101221_AAAWKO miller_m_Page_094.tif
288991d80103a8c8cd29a6aa9f833265
3a51c21fb644f754591413d139544a5b13532883
F20101221_AAAWJZ miller_m_Page_065.tif
637a57019b4602e2cb461b61d827c31d
c2391ef136358ce29bd85ef8054ab322d33653c0
F20101221_AAAWLD miller_m_Page_124.tif
37e930e73fe97355ccfbdd85741eef91
5f7385cb2515c1d10dc6d00612ac009e33ff37d7
F20101221_AAAWKP miller_m_Page_096.tif
785dc05128d40f0e0e675e7749f9e9ed
8d74423a468775182522bf05132556e4653d545b
F20101221_AAAWLE miller_m_Page_126.tif
82a78e8fa83ded289d277a8b8ca0629a
5d51bc58783d78f0ba6eca5bb89b55fb39338ff1
F20101221_AAAWKQ miller_m_Page_097.tif
b5d967c950b537cbcd8e5cd705cf7500
352219a8ecd75e60ac488f93460397725d2d7ef8
53964 F20101221_AAAWLF miller_m_Page_003.pro
22fcbab6939e8516a5887e23dfaf6099
a5cb3fc2ea9c757e22912e8e3c5f73539ba8cb5d
F20101221_AAAWKR miller_m_Page_100.tif
5bea3c55e76094428dd728080840128e
aed1473c6b1f41dae95c42f568420968117e7ec9
9365 F20101221_AAAWLG miller_m_Page_004.pro
b8cf3ac50468b8b033e157e7df38c15d
3b260b160f3bf7f3e1cee853de8a78999bf7b784
F20101221_AAAWKS miller_m_Page_102.tif
6a466986ab5021e9bb1aeb27d077d817
beb7281d459b2f7c68abaae0f5cf115ba3d9512a
78234 F20101221_AAAWLH miller_m_Page_005.pro
0a7e522ede29e14423fd73bfaf170a17
2e1f633b5206be6755e17c9935875c5edbfcd818
F20101221_AAAWKT miller_m_Page_103.tif
3863ccb1b00dbcd32ade45e8793a4d46
830f98ab03eb77d6b551637237c7bc75b29347cd
55902 F20101221_AAAWLI miller_m_Page_006.pro
d3539b13525475e058ba75544f5cedb1
ee72c13393fd7bf57abe009c7e1f5b437f4e3c44
F20101221_AAAWKU miller_m_Page_106.tif
b0346deb421087471f59e141b48653ed
e6be89596977832ccc89984d68a0f71f2f4098c9
33456 F20101221_AAAWLJ miller_m_Page_007.pro
888fc980f2176644184a6d511edad299
6216915517cc8c6be7363c901a09ee4ed742b468
F20101221_AAAWKV miller_m_Page_109.tif
fb43669676e93bf90506f0f0c0e37032
59af933b0f5f9a2e0f03bec4ca2fac1a72a21c5f
55242 F20101221_AAAWLK miller_m_Page_012.pro
6079da2d82071a2145779ec05ad6349e
3e020c4578fd72ba5748396fad9abc9284bf3b10
F20101221_AAAWKW miller_m_Page_110.tif
c04c7757253c0242e573add649108bcf
98f44af5fd6460d2659ee41cec4e08353baf4bc0
55465 F20101221_AAAWLL miller_m_Page_014.pro
7bbef1501c79702a7d5a47c873b99353
bebf12d2409ec4c0f9d19f949b34913c5d7a1d77
F20101221_AAAWKX miller_m_Page_111.tif
40224afc27f62aaf4b54d371df79d7cf
6c447663e6fae15ce9c4494f6c2440725bb0c064
55690 F20101221_AAAWLM miller_m_Page_015.pro
a491f494c39964f777313a0f6e564bea
65231af7a76385af8f785294c26b4cdb561ce6e0
62715 F20101221_AAAWMA miller_m_Page_038.pro
b6ea6159f389a7c0a6d3bbd447ba3cab
179bc647fc60f1bc8d7bf264ffe0db5e21bc5993
F20101221_AAAWKY miller_m_Page_113.tif
a60f05dddc6e0aa74a22355642833e30
958487bd28b75d7f431d119748b369aa874cd2a0
57444 F20101221_AAAWMB miller_m_Page_039.pro
634dda80356b6fe2cd460859dcc7337d
2b977baf02f1c4b98fccf7b7021f7cfde7182f3e
56446 F20101221_AAAWLN miller_m_Page_017.pro
d6a24a88d03f323b7df0885eff42c7fe
ff22622cef139750c3c59ca24a393808acf73100
F20101221_AAAWKZ miller_m_Page_114.tif
7a7f78691c3aedb63bbdaaeb19358ed4
c4a9ff4e08c366078a4ed0d298c15dc1d4efb747
56667 F20101221_AAAWMC miller_m_Page_040.pro
84434dbd7ee7b55ca4f3be14438d18de
c7dc779f707126a5ac2d82db474ab39c7a1eb686
55935 F20101221_AAAWLO miller_m_Page_019.pro
8e1d30c21d45777233b776bd98db77c3
55202c701f660d459a4d8956f1cef38a1a719f4e
18614 F20101221_AAAWMD miller_m_Page_041.pro
d6705f5cfa40de5c7fa6763aedda24fe
930e3acb9a6733b03132967870c5f7c8be6d2507
52427 F20101221_AAAWLP miller_m_Page_020.pro
c2404b8cff131a3700a348115cd9f7de
d54d1a48479d4c7a3fcfebf744b20d23fffda6fc
47818 F20101221_AAAWME miller_m_Page_042.pro
f48c462f6719eded3e5b65004c72ca6e
0126bcc9eac789d71c72f692c41ebd9ef99f19f6
56678 F20101221_AAAWLQ miller_m_Page_023.pro
27f7c1a72bf14bd1774c7affb8f4941f
44ce5dee51673c62fe1044411cc8fe35a7d38fd6
48031 F20101221_AAAWMF miller_m_Page_043.pro
c5dfb801c0b94dfe2b78a60d07510036
c3625829f82ca352a57d29e4b793aa1e0d136dd2
54569 F20101221_AAAWLR miller_m_Page_025.pro
e04236078261453181a194fedf5fb870
9cab6f7806c1c9542127d981737b05a4bb826c05
54714 F20101221_AAAWMG miller_m_Page_045.pro
406b40c996bba3f8e71cd8eb6f821552
ab6ae441c5ab2f5e1e585dd35f1a95c96a431d25
56841 F20101221_AAAWLS miller_m_Page_027.pro
5a16477d90390dfc46b886c3df063b4e
5e280640986da472e470a7ae380cc996e4e8a58e
57174 F20101221_AAAWMH miller_m_Page_046.pro
831e5210b35c79a89bc6a3c8a9165238
06b2a07343615d6a05b36ff50fd87d38c5a22ebf
54079 F20101221_AAAWLT miller_m_Page_029.pro
23c76687a9413f99423311fb48bcdba6
8f79b59e883fa01e6a2ecf444db9bc3ba4b162d2
48602 F20101221_AAAWMI miller_m_Page_049.pro
3cc590944a8f8aaf24471a9954241132
23bcc338c0b6212b590ad12f0c2e8f54558be5b8
56275 F20101221_AAAWLU miller_m_Page_030.pro
28f99efc6292b6ffe7df80078c4f8814
9468e2544e9dbbfd68aeb30778b416ec4852cccc
55295 F20101221_AAAWMJ miller_m_Page_052.pro
e1f267099936d3eb073fda1c3e13ef78
2dc28bc5312514d6c7eeabdd56dc010a450e383c
57978 F20101221_AAAWLV miller_m_Page_031.pro
e98190c1914467e3bf1bab4b28d4bd0f
a6d2c1becbc16709e83bf360cdec9aac2f0c1a64
56233 F20101221_AAAWMK miller_m_Page_053.pro
14d5a191a6203ed0b1b5651fc1210368
221a13f12f718315f56808e3fd063b8f9ecfdb8d
53630 F20101221_AAAWLW miller_m_Page_032.pro
5d979ce49e1c317112206641c8dfdf95
9399736045354bc5324f2bb186bbda0789c6d25a
53193 F20101221_AAAWML miller_m_Page_054.pro
d412bb81c77a4762303df91160eba04b
1518e56a9fae8753847141469e67d97c35582783
56982 F20101221_AAAWLX miller_m_Page_033.pro
eb7fa5832dbac930ea1b43cdfb96c010
e54ff01adba15332e20997a61e363bf957b086f5
57407 F20101221_AAAWNA miller_m_Page_094.pro
5a8757a6cbee32ca8471fdfa81ccea5f
cb550a084022002ddc9646adaf5a6e60abb8ff11
50929 F20101221_AAAWMM miller_m_Page_057.pro
71f5de6ab00b68cb590af7fa0607fd7d
885ba10ff5c800a2090722cdc194e986ed2bd247
55996 F20101221_AAAWLY miller_m_Page_035.pro
bddfa9f9f9168c2b948bb188edb05996
2b1f637f406b95b97eae0639fbad527a21406d39
57774 F20101221_AAAWNB miller_m_Page_096.pro
19cd2430c93122e9e4fde5b527f098d1
cba99caa55f74c055441b9fb23ff29dcded90a1e
55168 F20101221_AAAWMN miller_m_Page_062.pro
c81e9c4a211ba60adcfa36f2405cdff1
987e321b029d4ef93d6fc179d8895aba119a39f7
55542 F20101221_AAAWNC miller_m_Page_097.pro
78458be709f426fc0bc4dff48a72ae05
5a62a75372acd7a7a4c3ef7914d40f044cbf1fff
57045 F20101221_AAAWMO miller_m_Page_063.pro
a51d708168d6c608737c9ec213427cb4
9cc95e175573b776d2972d56d72f1b9d4285f876
57987 F20101221_AAAWLZ miller_m_Page_036.pro
77962e94edcaac4f3b135b665b9bd216
ab750dbbd63fbed53b223f20383c0c16aac7c05f
55194 F20101221_AAAWND miller_m_Page_099.pro
478f6d2fbe127b2b8322b6b021897ee0
03f6f31fb6ee781ce5f754bfd93c32239ceb3cad
53064 F20101221_AAAWMP miller_m_Page_064.pro
158120b730beb78319d3c6af90dbcd6e
7726cbb7035979570146f01e5f9043fd980f8ecd
56578 F20101221_AAAWNE miller_m_Page_102.pro
028fe0d4c421f015c3354a8da36d562d
340e9f1d511c4c57eb2efe0543bb464b9fa86ee9
48281 F20101221_AAAWMQ miller_m_Page_068.pro
52bac898766b0832a668e42d6debadbd
05ef7b6bad81cf2aba93682a916b90a9f15610fe
55699 F20101221_AAAWNF miller_m_Page_104.pro
c5c6c002d4d099edb993c44ce0d5c316
bffe988a892b9b708ab93b32dfcfb20a3a4a1521
53102 F20101221_AAAWMR miller_m_Page_070.pro
fe9f21f9ce541408635cd6f49c45cfc9
e4d20ac2fdb2e116a487cb9e5e16f2f80c818cdf
5228 F20101221_AAAWNG miller_m_Page_105.pro
ed89e2299c2a7e0f5d51b27d4e62a188
b6220bb788db2306d35da224998c496d7ddefa66
56075 F20101221_AAAWMS miller_m_Page_072.pro
bac843eea8507c6cc295e3c62f3087dc
cd8fd1911fb6e84e9781d0ac84d389bc63242010
68018 F20101221_AAAWNH miller_m_Page_106.pro
929f1fe18285ae913bd079e1dadefd46
ad95b83ab84808d88e9e48e4c82aa03eb705b16c
30621 F20101221_AAAWMT miller_m_Page_074.pro
643e7ddb0861112d36ed6be763eaceaf
51a5027fdfb53c6063f299fce14a184e33c720e7
32548 F20101221_AAAWNI miller_m_Page_108.pro
80a509d6ad48f53af4c26f1370af6db7
cd6c13fe53f876c7e012658ba90f0db537b5746a
46186 F20101221_AAAWMU miller_m_Page_079.pro
a236240562bad168a3325fcaed992499
76c03f3db2af5b6dc8848d8b0f5af272a41e7f06
78767 F20101221_AAAWNJ miller_m_Page_110.pro
6e97d481b05dcdb50285be79c7ba8afc
ba909a37fcaedcb1ec1aa0e23a1fdb19a7a49828
13502 F20101221_AAAWMV miller_m_Page_081.pro
3881060aff6761abbac546bcca086d10
a1c6d6d80a1b1a97ed2489f64eada2cb3f6cb1eb
75700 F20101221_AAAWNK miller_m_Page_113.pro
052e2b668d37b8c86a1c68576e820c9f
f2f096fa6644efdaab30d651826e18938f5bc9ca
43498 F20101221_AAAWMW miller_m_Page_083.pro
d45291d19bfb9793d559095e167344f6
b6ac11a2443cf7893e3174585874e146e4303dc2
45735 F20101221_AAAWNL miller_m_Page_118.pro
5c1a7a415f52a850792e228e59dff0b9
d23286c2a79f988302fd1fcd2322f4b7c7e81816
19860 F20101221_AAAWMX miller_m_Page_085.pro
f0a33d7f66e009aee3aa72a89b3ab2c8
b27128e96dbf888c0878f264d6d67ca9a76d2b1a
68862 F20101221_AAAWNM miller_m_Page_123.pro
956518541c7ad29224498ebe49a704ce
e5f7f6d4abb65ab7edda74db1336533857f20105
52109 F20101221_AAAWMY miller_m_Page_087.pro
7d3f89070da27ecd8f3246d16b955df1
8b5664211ec65ec3b84fb79c5111d70acb7c6061
2284 F20101221_AAAWOA miller_m_Page_024.txt
5c6e60f6f6a35ac491f1b8f91af85b10
f27e9dcb71186d22d98bce24893e79043e4f18e9
60398 F20101221_AAAWNN miller_m_Page_124.pro
f49e088bd5b3fec94e95ea77baa55da4
87e265e3b36f59c65771c5e13efbd55d6bc2259e
54326 F20101221_AAAWMZ miller_m_Page_092.pro
a4aefb25e7b22b8a46a9dacca473c99b
c0657a58542571d2b32b4902b65ea2f2a6dea7f1
2165 F20101221_AAAWOB miller_m_Page_026.txt
dde0700fcf1e481267fda113a37a99bd
d85ca248f80fcb447b2cda28043be1cd11743d3e
22134 F20101221_AAAWNO miller_m_Page_125.pro
ad2c5cd878b3bb34dd827264c1154968
7e1a542944f7bc15cd9b72dd4156a1cce628b5d3
F20101221_AAAWOC miller_m_Page_028.txt
5ef06461e58f6d7f9f013fd5eacb59db
87b35d947dbc85934fbb837302d39b244a0fa8f2
382 F20101221_AAAWNP miller_m_Page_004.txt
473bcfaa6d276a7c33cb4bb58d342242
32de7fc4591e59906f3b1166957e124b735eef26
2310 F20101221_AAAWOD miller_m_Page_031.txt
c2ffb9fb4e43e6eda8c8c1a5581eb7ac
b195f3f940b840a177f92d0fec4bf0b18b93e481
586 F20101221_AAAWNQ miller_m_Page_008.txt
cf8cc14fff20dc5a7475cdcaf3448473
297169c9f7959ddfd4ae540bc5e3cb8cca15fe78
2105 F20101221_AAAWOE miller_m_Page_032.txt
dbab82ba679b2790f239ffccb84f140a
e4a3a0836ced3aa78c4b8f48e03f4ae604d58f8b
F20101221_AAAWNR miller_m_Page_009.txt
bb79b6fe36acd96ddc39c77675e9e9e1
add4defc37f612de32530caf1570071a005f037e
2278 F20101221_AAAWOF miller_m_Page_033.txt
098b06bfcff343a799f1d15e7919058c
9fc3152ff74d1d00670d746de7c945e4e22c582c
2038 F20101221_AAAWNS miller_m_Page_011.txt
765458a5e32b9829cd94943460adc437
afa16ba34509f8910e7f87da3020186444254386
2217 F20101221_AAAWOG miller_m_Page_035.txt
fd0e11d0c0a9cdaf07ba54e807cd462c
f2050119652053e9d0f42e34af937edf6094125f
2211 F20101221_AAAWNT miller_m_Page_012.txt
5bf79bf077a0e4675bf35aa6014caa45
727642292e240db81810049273939c62ba3d0c6f
3061 F20101221_AAAWOH miller_m_Page_037.txt
116709adcc25ce0f472c25cad4ea2826
cc65ccfc3c05cbb061f8b5977cd849416b68d544
2185 F20101221_AAAWNU miller_m_Page_015.txt
dfd1f8a3047cc5f7b1395b9b17fe3898
42367c77010039e81cddde39dbad60b131f69d72
2523 F20101221_AAAWOI miller_m_Page_038.txt
e0a04712d3cacde8abd8db96bd8e0b7f
09202eb380c1fc64255e51d2c45e878f92e6b836
2213 F20101221_AAAWNV miller_m_Page_017.txt
4f336d5fdb55465561efb624a8453fe7
0a90dc33cad836144387871e3b698529a2abea34
739 F20101221_AAAWOJ miller_m_Page_041.txt
f338d36a03f0fe0dc63826435fe5fc6e
f58ca5361ab3e8e8713fbbca673259f57d5f60ec
2229 F20101221_AAAWNW miller_m_Page_019.txt
db1a5bf05b5738027724e730196c1519
50de9e8bfc1e3325287a567ddc52535398bc096b
2046 F20101221_AAAWOK miller_m_Page_042.txt
bf039fdb1580a3bc9598530074e5a608
b6a7f40789f3d2b355acc368ece102f49111b3d9
2095 F20101221_AAAWNX miller_m_Page_020.txt
1ac6071f9ca8505cbc16eea5e957d584
67ae1327926c793df59819342bef63734b524411
2014 F20101221_AAAWOL miller_m_Page_043.txt
5e9b61d7d945e42b7c72c8a1c8edea63
ab54565486d8e604790b2cace3c5f4d75fa9aa27
2114 F20101221_AAAWNY miller_m_Page_022.txt
97cf56432f605da95ceb36d84cdcbbc0
6e0ba268925df7ed529b6b2c0d3b18bd2fb6d3c8
2233 F20101221_AAAWPA miller_m_Page_071.txt
4a7ebb86b6ff097347a84ef73183dc01
71d5f9d9c5c32c52521a97c8622159336e3ad6a1
2148 F20101221_AAAWOM miller_m_Page_045.txt
b50df1690519af8d9f78941fbb2bfee8
f17ace3dfec02437fb15a729383cdc3e76fc0b27
F20101221_AAAWNZ miller_m_Page_023.txt
879563033a8f34a8eac1ecf2310d33ba
e46c2501902544e30a255b6e848f59c903a881c7
732 F20101221_AAAWPB miller_m_Page_075.txt
17ec684b4c97252d5a384771d551e665
4438f69ea679f6dfe761e7ac2519e5b1f3d2ac94
2035 F20101221_AAAWON miller_m_Page_049.txt
68865879cf3e1d4c020108a108f6ca17
949da8e92090b636a8f0d7f08ea5a2b3a40d1bd6
1104 F20101221_AAAWPC miller_m_Page_077.txt
2caaeba5cb5f9693da663165eb6551fc
e4dc4f897809fa940f7913733d9ec9c113409e1c
2184 F20101221_AAAWOO miller_m_Page_050.txt
3ca25ab56f0ce0b14f07e8e4f2e708ee
05e3887cb0dc8120a1a12ad10fe39378018ce7e5
1870 F20101221_AAAWPD miller_m_Page_079.txt
c78c36a191b9342fbc95e239a7e5bf12
5c2f3079171166d917d9493a5c696b6c8f951da1
F20101221_AAAWOP miller_m_Page_053.txt
e609e2d3efeb21439453f82334910e5c
613e8f12e78db9f35b472aa1800d2bdf6a7021e0
630 F20101221_AAAWPE miller_m_Page_081.txt
a6a56271a3d128225fbff5eac77f01a7
2d45e8452a2e83879343efbc80c69636246e76a6
F20101221_AAAWOQ miller_m_Page_056.txt
fa6d8b2a07880ed39a50bd26f00cde47
a855d1c94aa5287ff13770d5a8610e38bdddb34e
1406 F20101221_AAAWPF miller_m_Page_082.txt
3a164cf68d1d6c274ce420b399dfaa28
f4c23a53edb4ac5a28ad33745d30f3a583ddc390
2261 F20101221_AAAWOR miller_m_Page_058.txt
07616bf14fb42d416ec4f059b72cc985
938f63e8de02856519a9d05c68fdea2d9107ca5d
2153 F20101221_AAAWPG miller_m_Page_083.txt
beda06c734ff4b3c18440cd1b435f9a1
80f7604be9966315d845e32392d34e8ff749119b
2104 F20101221_AAAWOS miller_m_Page_060.txt
624101b9c2c7eae49af4c20e13918faa
0d3eb168bd7f5e6b93789c91d950803c9e839371
951 F20101221_AAAWPH miller_m_Page_086.txt
56dc60a3213a1fee0ca78ae84ee70b1e
685d5d781ae278d4ee7a3477c73e5973183b11c3
2084 F20101221_AAAWOT miller_m_Page_061.txt
3f00b5d222d0eee085a85b71b812b2cc
a4586f42cfb4af356f7a51456bfb040e45861599
2130 F20101221_AAAWPI miller_m_Page_087.txt
2406562aa4d6fe49cfdc553c7dd2f69b
9da7e959399ace9547fbd9b70435074030a73b42
2162 F20101221_AAAWOU miller_m_Page_062.txt
0d9030b8af2ca6a000eddc39c2ba67b7
aff2285d1d7a4b500c4d515983fe9f1b03dec161
2256 F20101221_AAAWPJ miller_m_Page_089.txt
febe6f723020af8d03becf1477544ab5
9ecf2753fc22bfea916411564cdbed62c7b10750
2235 F20101221_AAAWOV miller_m_Page_063.txt
5c5a18b7dd7b28cf8d232771064c115d
50c5e81fbe631eebb53614613d95009ec1c2b259
2206 F20101221_AAAWPK miller_m_Page_091.txt
94330f941fac625b42d247f106a24bf0
a1826550e2b068e7756334cb1c48c3853889b1ba
2179 F20101221_AAAWOW miller_m_Page_066.txt
b2ac8c6b94f1194003c8730c48ec74ae
df1296273c1b766100612b79147c48c40712ee6d
2133 F20101221_AAAWPL miller_m_Page_092.txt
4a8f2d2e15b817db694e18464a21fa56
ff1873fc4520a7fcbdf749872b760f3ef5c9e7a2
2156 F20101221_AAAWOX miller_m_Page_067.txt
bf57787f4e73484079291bca992395b4
4d785cc889f20710acf0c750013504dac91deb2e
2215 F20101221_AAAWQA miller_m_Page_116.txt
75413807817f6ee1365b967c1a54f6b5
a5142adacf2c45b9704782b588ceefe99bdf2816
2251 F20101221_AAAWPM miller_m_Page_093.txt
022baeee6eae432d76167a02b8b48579
cf69ec38323f35c62882b049e28bb909cc2868f5
F20101221_AAAWOY miller_m_Page_068.txt
1343af4bd490ba3b20811161c77248e7
daf493abad5986d0d54b531f7218eceb7860622e
F20101221_AAAWQB miller_m_Page_118.txt
e6912fd17991d6dcd3126e32905e102c
4b2023448a4c663de02050191382e4c19d54185f
2277 F20101221_AAAWPN miller_m_Page_094.txt
ec5fdfdee99d9e9f25e2d5c8a1809a68
bb2ce2a8053422e34e986d5a32b430c8be64a9fc
2170 F20101221_AAAWOZ miller_m_Page_070.txt
26a73b88d2df2dfe3b95f85bb99c4042
cdbb4f44e84375a941142a685b91eba5ab99da05
1097 F20101221_AAAWQC miller_m_Page_120.txt
9b6968522dd9b45af083155def783d7a
778f8be03c30230fdc461960157ef3ee4981812e
2182 F20101221_AAAWPO miller_m_Page_097.txt
5a8b3f0bd4dcace8a2977168b819d77f
7986084ee45f8adc754460290edb16bdaf54829f
371644 F20101221_AAAVNA miller_m_Page_119.jp2
01bd44edb1e279e6f02cfb54f9444d5d
c5a0e20d6daa1f87d66bf8ff2f8794696679133a
1061 F20101221_AAAWQD miller_m_Page_121.txt
2096b3846eb3971aff2851fc596d0c53
6d7279c6ab61d8394d674d287d1dfb5d53045c4d
2161 F20101221_AAAWPP miller_m_Page_099.txt
7213e1aa6bd2f7ad005901a6ffece9a8
c33fe2712aeba386e17afd6fc71c3a074d6de8e7
10791 F20101221_AAAVNB miller_m_Page_111.QC.jpg
b5588c1dcbf4d3a84b5cd986a0df192a
f02a556f4cfc09aeaf96f71d90b2cf4ebc661699
2488 F20101221_AAAWQE miller_m_Page_124.txt
dd1a1e473f2888deb140c7a380311733
ebf1947e09c15a6316293dfaea83a9c5860ba273
2142 F20101221_AAAWPQ miller_m_Page_100.txt
ff036b33ac348134fea436b7dbddb677
535a064e5f849338fc39259536cfbc7f6acb246c
38026 F20101221_AAAVNC miller_m_Page_036.QC.jpg
e23962eaa3a3700c3430c2f89b767632
2fb0384deab653c99be55d4ba8bca5d3ccdb0d8a
622 F20101221_AAAWQF miller_m_Page_126.txt
14f9a0848a66201097e5a008cb1e05e9
d62c6c926fdf76ec9eabd113336113925dd3f65a
2166 F20101221_AAAWPR miller_m_Page_101.txt
7a920680e9711035684c75bd0e431644
86f966a30d654d0030b7b92d7dc02415824da55d
8312 F20101221_AAAVND miller_m_Page_087thm.jpg
2be281c093e30dc1ff962002c9c09c67
534e2e6ed58a2480e7a2b2d223f8ca2ed67308d7
34662 F20101221_AAAWQG miller_m_Page_020.QC.jpg
8690df2a67b9bfbdd1e1358338a50c53
a1b33e8ff28d7b44e9298033d357feeec5e22fa0
2223 F20101221_AAAWPS miller_m_Page_103.txt
dbd42d091a556068d9baf64f3f46d482
906d3a86bf34bb1e1452633b6d62f2b8cb2c2da9
15981 F20101221_AAAVNE miller_m_Page_111.pro
4a010166ebe0ff99c0a2be1027870e31
16b7046297ce08a0732cd896a194242c166f4234
9017 F20101221_AAAWQH miller_m_Page_055thm.jpg
80d37e42a6e70a18d943d0f50536b10c
ff66d39758ecb502f946080c9fb659d71487daa4
2183 F20101221_AAAWPT miller_m_Page_104.txt
109e6a454dff59d897948821c785d278
b53b30a7d77a5459fc375c13c129e05d3f05fd87
36598 F20101221_AAAVNF miller_m_Page_059.QC.jpg
881202649f790daa1aee3628e4ac95c0
a8148667b2494c736a264095db54bf9363ad9d1c
19308 F20101221_AAAWQI miller_m_Page_109.QC.jpg
6013b1e28fbc97a37ae63a1af59fe11f
b1c90ed26e69352162fe454321022c9eb8e4c0ac
2822 F20101221_AAAWPU miller_m_Page_106.txt
6c26be470ae430177bc7c4ac420722d1
8a0128ae42a64753062a65438b4748026cea95fe
135451 F20101221_AAAVNG miller_m_Page_114.jpg
f02cdf9a2c10a66249e6e8424557da96
88e49ff80202eb8a89fe7b2983c69349404b9d3c
36074 F20101221_AAAWQJ miller_m_Page_034.QC.jpg
a6b5e49ca027ec486eccc7c45c37f431
bb8d5578508d65d55c8174aa5d8219f92d2a01b8
F20101221_AAAVMT miller_m_Page_072.jp2
eeacc6705dfae3d36d74e3c628f6c00a
468baef69887c463c506f8eef861eccf0f9014f3
3155 F20101221_AAAWPV miller_m_Page_110.txt
c8bd59c718012b0a4897db28d5015825
922d22a157d3046f91ee526c3bb0ba259377ddfe
2288 F20101221_AAAVNH miller_m_Page_006.txt
0199f3f7552d8d5eb6f83e75efa41676
5a00b83da54ea3fdabcab89d8d8d9d1f356288b5
109704 F20101221_AAAVMU miller_m_Page_045.jpg
ecc769dc034a1439986c379da38ae840
afab3111684056a3740787caa8e8ccd2ad5ca9d8
706 F20101221_AAAWPW miller_m_Page_111.txt
74c4da4f6f14d607e155499a729257d1
a7d46e69d2472109442010fe33796b75e119d947
9009 F20101221_AAAVNI miller_m_Page_104thm.jpg
3b7369db71a97f13346eb2862fc9f220
1727445545e3d7dfeed799dab001b37bdccb8eb7
37527 F20101221_AAAWQK miller_m_Page_089.QC.jpg
e7f0ad7d16fddeb9c8005801d8f4856d
dae2565058d0268842701f5454966a80c75c5be5
F20101221_AAAVMV miller_m_Page_007.jp2
38c8bc53914052ccb38f853dcf123ae3
7f2d415483c46ee61e1236acb9517e128fce5ca2
1680 F20101221_AAAWPX miller_m_Page_112.txt
636a4463c3f9a2eae63aac0ba09fece8
c8ca82847b04bbc11ddf22c65d383bc5fa2c6399
36757 F20101221_AAAVNJ miller_m_Page_027.QC.jpg
00e97e595f4bdd3b3338c0b0014d048a
3a660bbe040ced18433a1916e4c4c5c2a6f51ad2
33880 F20101221_AAAWQL miller_m_Page_057.QC.jpg
72e7f2e2da243dbb268557e68907890d
f2ab4e65859375fa5123796536836dfbfe2ea5a2
4398 F20101221_AAAWRA miller_m_Page_048thm.jpg
93424496ac2f14cf957163fbd938b785
ea6080e06b3c411363ca1edcedfc05a4c0d411ed
F20101221_AAAVMW miller_m_Page_031.jp2
df18e4f8fe52ae652e68c8704d168ab9
4b2a7bacdd794cb8f0966ddeebba5805914506dd
3357 F20101221_AAAWPY miller_m_Page_113.txt
21cec287fb3ca00f33bd935b0c478a66
1c012f460641e80134350d18fabb07e5607a3418
8499 F20101221_AAAVNK miller_m_Page_039thm.jpg
39a386e6709049d720150c8a8e27da87
21e8820d606f26430131f7f1ae0c7522c997e0bb
2589 F20101221_AAAWQM miller_m_Page_081thm.jpg
4c0aabeadebbffda1acc1db9945a9233
d232f2e3193189a1f71966420685a357f54f7626
35173 F20101221_AAAWRB miller_m_Page_088.QC.jpg
0ec1016b6cb63c7e21cf3f8ebeefd724
2d66a123527a8f8b5b66125ed6f398971c6e8fc3
8611 F20101221_AAAVMX miller_m_Page_020thm.jpg
04d3c76f3b06d2672a8e402501a22541
b0ec1955efb656a5ac01228246162f6492883912
1134 F20101221_AAAWPZ miller_m_Page_115.txt
82ef9bc968aa2c3dc8a3bd7372d1076d
cea799f329db77bf4c3bdb43ae3e17f883edf477
506248 F20101221_AAAVNL miller_m_Page_125.jp2
9393df83dc470012490307f963e52305
72b4a62849533b2f75f1bb2da588fbf8de0b7b00
9201 F20101221_AAAWQN miller_m_Page_024thm.jpg
bb6d06541545f5320e436ddbc53bfe13
351a9258962c15f0157a81802642ddf632217807
35410 F20101221_AAAWRC miller_m_Page_065.QC.jpg
2f46a1c9960c9c9344e4b691f37944f3
af27dfadb5425a2575cf6fb7a639a56198bb8441
F20101221_AAAVMY miller_m_Page_107.tif
56c1223109f2fc39d39377ffe2d87305
7f353c2042bd722e4d7d11a470bb1cad8f52b25c
2230 F20101221_AAAVOA miller_m_Page_027.txt
0ae9f5689108db7b290a7da841631538
9ef1e732a7913059d8bca135b223a71ecd28681c
28993 F20101221_AAAVNM miller_m_Page_082.pro
0b0a23a72b88b86a72ca01ffad513df6
88aa29ccfe62b7338b161de8e15bc11559eb9764
28048 F20101221_AAAWQO miller_m_Page_006.QC.jpg
7437197f34298b0dbd68f82533779aee
9b34c7acfb0255d9f508f3b51de97b6bbeb167df
36291 F20101221_AAAWRD miller_m_Page_062.QC.jpg
151c5c329fcd0712da21b2a03ed482b4
1643f87820a9d2f0171403d1d7adf25083d554ac
8693 F20101221_AAAVMZ miller_m_Page_022thm.jpg
b70f0736d9d18bb2e23036cc76f5a319
5599633123f004dda15c0abaaf7c78966b6d981b
8997 F20101221_AAAVOB miller_m_Page_099thm.jpg
0d047a27b2126a045013981edbce5d5b
0d5902282bc22bbce3147d8722bf560583dfd1e6
130813 F20101221_AAAVNN miller_m_Page_110.jpg
a8dfa0fd21ae4df68c050a1aeffba969
f06451797e62801e8815082b360b435573ce61ed
36477 F20101221_AAAWQP miller_m_Page_017.QC.jpg
23cb3f5f27fcd5b7377166d5e18addcf
d5874c0e8754f7b90efbd2d6afdb5c8f1ee05289
24810 F20101221_AAAWRE miller_m_Page_083.QC.jpg
b43ce92b490cbd1418e682aca46974f5
9f3fe4d62ecb7df3dd240dc63838b5065542e8b4
8661 F20101221_AAAVOC miller_m_Page_066thm.jpg
ce2221e2177c06042516c60d3979283a
e7a045f739a0c2fbf30b5e00ce87bf9919e85b8d
F20101221_AAAVNO miller_m_Page_020.jp2
7c9de352028700df7d5cf2fa87c9704a
ffa2745b24394497aa55be58e7244c975b6fab53
38447 F20101221_AAAWQQ miller_m_Page_013.QC.jpg
72b79f3777665f652dcb5acfa5d041a0
2a818263b322f9f79b481703e51ea6fdf29bf18f
35988 F20101221_AAAWRF miller_m_Page_054.QC.jpg
74bf8e72b7b0205036ba9c6780efdee6
663738a27290968a5d77a99d279607f86bdbdab1
112345 F20101221_AAAVOD miller_m_Page_067.jpg
f57c92e5523f0d6ed5411ee949b1c18a
a5a9a402f9cb021d5f6d0cef6270fc24c59bbdc6
56133 F20101221_AAAVNP miller_m_Page_091.pro
b848c2bf1d20037b5a1a3516cff703fb
9f95b16995ca38cbb0656dea718761bfc3a80fd9
8908 F20101221_AAAWQR miller_m_Page_037thm.jpg
0849ab8aed9cf3a71c108a3fd44a7e8d
dcde54902cf81b5b88798153303fe32ba8964bb7
4671 F20101221_AAAWRG miller_m_Page_108thm.jpg
9c383e29b600ba0cacd445d1ffeaf097
fafd988d96956d836bdbc1a0a110d0ecb2a64622
36084 F20101221_AAAVOE miller_m_Page_097.QC.jpg
b3c54335b69c96028cd3fdcf29c67111
2d5bea54a4539c1f836244533853fcb950c70c72
55472 F20101221_AAAVNQ miller_m_Page_050.pro
609805009af49b6ed58e2f55c74225cf
8025c16f4c16a2e4702abcedc3dc159d18972d44
28641 F20101221_AAAWQS miller_m_Page_117.QC.jpg
f2772a8ad45b85ad33be272ea17a4167
62650986bf1daffd25c756a30d687f7caf50ad39
6516 F20101221_AAAWRH miller_m_Page_006thm.jpg
1119d92c2bedb4f01b57002aaa614897
c0c14bbbd23e508f3de07a9c9411fc4f10a90c5d
F20101221_AAAVOF miller_m_Page_028.tif
987dced7fe09daa0f70b4d9d68e14552
3c4b0b0050ba15552d614e1724984c2d2b12af83
2262 F20101221_AAAVNR miller_m_Page_051.txt
ee60f132bc3ea9a4a42287ab1d8d96f3
be8c3e11d67005632c6fe096ab572d8f4f3a4d58
8821 F20101221_AAAWQT miller_m_Page_033thm.jpg
4cf72351ddda9695e01058c454ad7736
af0402ba74409e1d097dc3f69faae75b6e8c1d52
38075 F20101221_AAAWRI miller_m_Page_069.QC.jpg
db0f6cbbd14fab844e7267f8c45ca9b0
0defc0c7d4463bc4eb0820ee75c2955627f1db15
1051919 F20101221_AAAVOG miller_m_Page_058.jp2
a2fd9025911c2e9ce5f1f89367f1f383
bd324f487fe0c657f35dd8fc2b9e15cdf1481b22
F20101221_AAAVNS miller_m_Page_067.tif
c2c16e56bd88fd3d2b09aeaf498256c0
76de864a314ca8a0c77f28ef9dad622895df6bbf
35670 F20101221_AAAWQU miller_m_Page_052.QC.jpg
b03a2ddcb76748369fb99b49eeba946d
d0100c27cd1dcec8357efd1ac677d89b167cb9c5
35401 F20101221_AAAWRJ miller_m_Page_028.QC.jpg
45e5ffcc4141621b59fc5acb512f097d
ea0b54ea50e55020a1f94c72dc3cc1f1b7731675
8985 F20101221_AAAVOH miller_m_Page_010.QC.jpg
0266eed3de55d2060542474fe57f61e1
48bc69f2426a3a7a281a1d8b67a431931f5279de
F20101221_AAAVNT miller_m_Page_065.txt
f9a8e2877e70b07e7622fd829bbf85c6
8c90c95d466d73135584b40d23887901c5f8299f
34055 F20101221_AAAWQV miller_m_Page_110.QC.jpg
43725388597424400a50a412cab144b1
3266190210ded69d600162437150b23e5d7b2c6c
17618 F20101221_AAAWRK miller_m_Page_108.QC.jpg
fb6fbabe1714b68f5e2c010145b07fc7
ef13e040b320c3458ac048831d985eebcc2d1e21
3781 F20101221_AAAVOI miller_m_Page_119thm.jpg
1bd4f7cd2e71f001e90ec46952ce0d54
c26be9fe1ffa3ba04cdbd8eec917902036c09a53
114719 F20101221_AAAVNU miller_m_Page_046.jpg
c1e8614eef608fc6901a44c0e770d851
d69fca26ba2c5be401b62a4a87297a711ccd4743
24595 F20101221_AAAWQW miller_m_Page_118.QC.jpg
0ebf3d8ee56a18bc11a5fd4529745a42
ed98f3fd9cd6a97110dd88513b991158fff54754
35317 F20101221_AAAWRL miller_m_Page_101.QC.jpg
412aed98810935e9f8e685efc43a801f
cf164a17f29e0808ecdbfe59b58ed943d23f5691
F20101221_AAAVOJ miller_m_Page_042.tif
5fa8ce82035926be8bb312db62b51c15
15ec602f12c1fe6a8a06d64d9852c66e4cf53a0c
4972 F20101221_AAAVNV miller_m_Page_074thm.jpg
0350508e2cb62226a70bc6422478ff3b
68254f9905d23077e1631750b8f4d6b8e2086bfa
8805 F20101221_AAAWQX miller_m_Page_058thm.jpg
28781c953c232c652f1f12c6e672fc04
8272f7a8f136cfd7c2baec573804922801c335e9
8796 F20101221_AAAWSA miller_m_Page_047thm.jpg
4f54d8c518899fc0802aa62d3ba05117
b3ff901b707121f8f9f5a582ecc232c1456d255f
8953 F20101221_AAAWRM miller_m_Page_059thm.jpg
b0f6073e681da2e93159fb4e305a991b
35cbbd94d68916949342b377e1e4d364eb193328
1219 F20101221_AAAVOK miller_m_Page_074.txt
03bce672d04a1e5a431a05b0c408d6cd
7d69d0967dc13e975f50b1a3aec72d69a5dd3bb7
8755 F20101221_AAAVNW miller_m_Page_012thm.jpg
d32f59f0c840a6875ee15308fb65b0f3
ca06c13a14892028aaaea5b4b2027cb1db321c2f
4882 F20101221_AAAWQY miller_m_Page_007thm.jpg
0d1a598eba18cb7c8fd9f339dbf6c10c
b941401e8869addc79318e7b86c7bb6478ac9abc
8711 F20101221_AAAWSB miller_m_Page_069thm.jpg
9d4d65e26270f9006ebb09c42d6d3e42
27105fb5b5d26a7eb60a0bd8eceda0274f737a60
35638 F20101221_AAAWRN miller_m_Page_104.QC.jpg
ebf6a031dcd416980c2b185ed0348298
a4c00186e343b859980855439983e79cece3c7a3
36035 F20101221_AAAWQZ miller_m_Page_014.QC.jpg
0066412461e9b32fb85e4e37b178bd32
b9ec10c7035863d1c723bb2a3238eb7f95992394
8929 F20101221_AAAVOL miller_m_Page_045thm.jpg
7b847395885fa7f3ac16edc269863aac
81096009ffc48ca5b909728f50aac0256a463f01
1051888 F20101221_AAAVNX miller_m_Page_029.jp2
d63261e9af396c1c20ca71d71998cfb8
da5fd352ac7503783db9c95c089d14a0f6200975
35361 F20101221_AAAWSC miller_m_Page_090.QC.jpg
26b019c5e773a89c07e6aa8e3cd4b3e4
c723a755dda310e86d215ebb714cc0b6e93b6fd9
35973 F20101221_AAAWRO miller_m_Page_099.QC.jpg
defff6c7557b98f05a71f7f2ff21093d
51aef38b33c1fd4563ef8b31edf44b81df392eff
2208 F20101221_AAAVPA miller_m_Page_034.txt
5230e0c0ddb73a10824d80abf69c6b8b
bdf1f195af153a17b5b5ae80d2e6fbb1a9505023
57394 F20101221_AAAVOM miller_m_Page_093.pro
a35ec8e80875536ee8c19527c7aa115f
ebacff66f61f0849de688d11560261fee631d67a
F20101221_AAAVNY miller_m_Page_037.tif
ae63dceed0dc81889630a1c0f480b7ef
5c1cad6965f4af2745a07819b46ab5b1bfa5bbdf
31931 F20101221_AAAWSD miller_m_Page_042.QC.jpg
3d7de20f8b96a88234f764ab892b7966
4cc04e63a42752dddfca59a69dfc52bc4dd9350a
35452 F20101221_AAAWRP miller_m_Page_058.QC.jpg
b48a2c1ecaa1af78c5cd2ad8b88d763f
537cc07f5d7970f0954dcbac447b5a09f1ccb02c
55050 F20101221_AAAVPB miller_m_Page_026.pro
2c9cfc95527c8c507e67b5960a635c6a
0bf3e5b1b00084a71b9eb5bec9b28ab965a40862
356457 F20101221_AAAVON miller_m_Page_126.jp2
999e7f89f5646210561b49c8c14ba694
987731d3dd9057597787a32d8ab7ee2fcd73616c
F20101221_AAAVNZ miller_m_Page_006.tif
7e2924f61b1dc80c122e864588a4b340
2ce2d0b43669d4cb77af5408c6f3f5914d430abe
2295 F20101221_AAAWSE miller_m_Page_010thm.jpg
4e05c3b05c310b318dcef3eb3363243e
5d21a2a4970f790b5246e7e07d563531327fc424
4118 F20101221_AAAWRQ miller_m_Page_120thm.jpg
b3d7887e04f7980a24e95ad99ae71254
147e615d4679e1cdc736e6d4ebc29db80d6936a8
2304 F20101221_AAAVPC miller_m_Page_096.txt
51ba0a7343d969fe98539ed3e4a022b9
ac6c6c1e5cf67b7026b8e88fcc5e4a85499ee3dc
F20101221_AAAVOO miller_m_Page_121.tif
b91009f9e9b7d3a3d0b6a6e5b67cf1cf
e5571c9056cc79a8a776466c1073a1208c4b4938
8718 F20101221_AAAWSF miller_m_Page_029thm.jpg
e3089cdfc15e9ed566b68789a5871834
1bd54db7bfc6c4228bc2bc70602da6b4cfa00c62
11481 F20101221_AAAWRR miller_m_Page_075.QC.jpg
6ab11bb522af703f161fdfd4425aa491
b8db84a21db16ee727e3087528fdccc9e3a9713a
24161 F20101221_AAAVPD miller_m_Page_120.pro
c825e30bf7885872cec43e997d62543d
835a0c360d5c95f0a059f84db5e7f59e00f1ae30
1051941 F20101221_AAAVOP miller_m_Page_053.jp2
6033dd8ddca5dff01df26a747ebd5484
beb701e06d9196de686ef3c6a0123a9146f9b343
3593 F20101221_AAAWSG miller_m_Page_125thm.jpg
ae092fed2738e6f73fabdaae68f9662a
f87833d7efb55726ffe21861ca976dd40f7b3c17
36484 F20101221_AAAWRS miller_m_Page_044.QC.jpg
15cdc45dea530e9631439ba50e94cbbe
c82269a7b5aa7d3567502f3497eae07ee107a194
112023 F20101221_AAAVPE miller_m_Page_053.jpg
dbbd192d81e80b22fc037233c276af87
de14dc124e5ca7c023e4cab8c79f3759085ba2f4
F20101221_AAAVOQ miller_m_Page_089.jp2
f5ef94e575106248bd70a3014be77660
1b2f57a8280beb0869b65a743817b44d828a9ed2
7915 F20101221_AAAWSH miller_m_Page_001.QC.jpg
77b43b3a37106b959dc233375251bd1a
f6b71e98d4d0fa9172b250f727da10d7c93de0c1
8756 F20101221_AAAWRT miller_m_Page_060thm.jpg
da86d86bfd1cc2cb5a2f4ae37501702b
dd1254b1bf6e8862f7e0780ba5565cc142447166
8599 F20101221_AAAVPF miller_m_Page_019thm.jpg
c960526246dd7110e06d0c6d85c51c9e
a48f2f0f8fad97a29ff6533e800b5e5611a9dc45
17205 F20101221_AAAVOR miller_m_Page_086.pro
4bfa3fc2c4acebd4a77ad10e3a169a79
1d1ac50c5df048e08563e27b641a371b9ffa75a1
8970 F20101221_AAAWSI miller_m_Page_052thm.jpg
77ec2fa8b7ca1acff1cf79d9ea8f3e31
aee22078bafe0404662edfca03220b60873c7f11
35465 F20101221_AAAWRU miller_m_Page_015.QC.jpg
62d543471c9e3766e37519d5a3b46713
9baee8809fe5ee5da38f902d40006a22e2c8d7f9
48574 F20101221_AAAVPG miller_m_Page_009.pro
48a0dd6529b8b58fa63c8a19bf235e26
89a0cd47ea654f3b4cafed58c46913a2f6dbcfb9
107823 F20101221_AAAVOS miller_m_Page_022.jpg
c11652f8eec01abbeae4fc6fd37bbdee
6bd3e306386f66b6a3f6273c7c0ba6506074aea3
3256 F20101221_AAAWSJ miller_m_Page_041thm.jpg
c280a3b88ab3c7b4fc241a7f1441e3ae
af0c49cd1982f67b62e672041d111705f50adcba
25000 F20101221_AAAWRV miller_m_Page_116.QC.jpg
709e0d4820c0a24e29fa5424a61df947
e6520019fd54c5d0d4a77b98371084712f17a91f
7595 F20101221_AAAVPH miller_m_Page_043thm.jpg
6c594d9ef1e69916448affb988627771
670b6ab24517804cb5537cb617f0798f868e5c53
F20101221_AAAVOT miller_m_Page_019.tif
94e47ef8b7360770933984e64b70175e
c82dfce88b343a071ca5bf6ea17ea41c2b54c41c
8517 F20101221_AAAWSK miller_m_Page_092thm.jpg
71255edcf1498fc422b506785d44ac5a
9e990a05b797e8d17510ba90a5923daaef040bd1
2758 F20101221_AAAWRW miller_m_Page_008thm.jpg
e86139ac2968b62fde59c300aa5919da
658a322e303da241a5d16ca281a30bbb13172940
8333 F20101221_AAAVPI miller_m_Page_113thm.jpg
446d243abb5cdd5cfc09f73b6463abef
19c6e7791a6667fa1a5ffe7b47ebc2400acbbf85
109847 F20101221_AAAVOU miller_m_Page_090.jpg
8fda20e154d1b0d3666e6468e35beef1
f503232ecc41b8d00030a3cd6f614b7b39a1c5dd
7782 F20101221_AAAWTA miller_m_Page_042thm.jpg
12e44d3ad56a2e16e56fd519ef82359c
0cdf51c3ca9c1c4d4ff3b931586b860a65be39fd
8941 F20101221_AAAWSL miller_m_Page_036thm.jpg
4b7034b866b283037380568ba105e4ed
4ff5589538fc9d57dc3d5197ea1fda6bcb91db11
36479 F20101221_AAAWRX miller_m_Page_055.QC.jpg
c55a0dd950945ebb3ce7978422f4f0b3
c89ffdd71e98cd0313217d1eaf528d15d2315eba
23664 F20101221_AAAVPJ miller_m_Page_115.pro
454e814150658a05f385e79c26daed9e
bc71fbcc12893fb5242b61994f1d42ce6ef2a98b
9025 F20101221_AAAVOV miller_m_Page_016thm.jpg
f9a9f03a3ba438911dcfa75ef9f79b5c
30c8c884a4bdc82a55348d557978f4cc9dc19338
F20101221_AAAWSM miller_m_Page_004thm.jpg
2d8bc55a5c7da161a60b7e8cab57f9c4
18831a4d0eb15c59b6899786589ebd0a83efdbd1
16318 F20101221_AAAWRY miller_m_Page_120.QC.jpg
2b9686b5ec8b75a63f8aa2a54682a083
e746d8d025cf306cc9c9a93f5ae065fe1556484e
36653 F20101221_AAAVPK miller_m_Page_094.QC.jpg
1f3c2388cc2e6b7b38c15e6893127514
e94fcaa2fdb06cdec7f8d9798c716fca87e41ce1
666106 F20101221_AAAVOW miller_m_Page_082.jp2
c4599550e766948ea50de05c314cd382
c3ecea28a39d099eadaa915d94f481e8e55b019b
8996 F20101221_AAAWTB miller_m_Page_050thm.jpg
72f4031e0fc58797a8bb8a0e7fac40ad
750cdee052d93596ea92f5e9156a0041fa29ddb5
10037 F20101221_AAAWSN miller_m_Page_084.QC.jpg
02131e6edeb9e901ddaf9aa5ac24a557
014b6961281d548932a6539a2a634d1c1b871c17
6282 F20101221_AAAWRZ miller_m_Page_112thm.jpg
8e5c06475e3a0a5a6ba5f6e8f78a585e
fc79cc778a9eb992c5fe4388f206c4f2ded50317
9018 F20101221_AAAVPL miller_m_Page_067thm.jpg
92602bd3a2aed4a034c5744a8778dd79
a6bf785808a4cfba15fa75fbb2dbea6738e5aefc
F20101221_AAAVOX miller_m_Page_076.tif
784bc226c2ff0fc70fd149ae483bd0f3
b92f159c6975377dc259ca505f97ccb470e503fb
8869 F20101221_AAAWTC miller_m_Page_003thm.jpg
365f0b894494e44199e111e3741da723
a36b2c8b77009518153f46ab62dade111b366b52
37127 F20101221_AAAWSO miller_m_Page_066.QC.jpg
0760ce01d9376f083c43a699711a27fc
49ee9970a0eda29d27f30505fc19372d5ec51b65
58828 F20101221_AAAVPM miller_m_Page_013.pro
efe872bce111208e5500157f400a182d
f4d6e9b13fb830ba27bf7736902bfa5ddb5ee0ef
941 F20101221_AAAVOY miller_m_Page_119.txt
d59c42cb4d45d8f3588fe4300931074f
2cb2c06aa8947726a7286b2f92e1f3146ba5d456
110628 F20101221_AAAVQA miller_m_Page_015.jpg
b47223f087d6aede44f50af66e384d77
86fb8a53cd5ec01fd7c6df7226ab1e2835c5f87b
11647 F20101221_AAAWTD miller_m_Page_082.QC.jpg
c5c31e8c403e30241bd360dd5ff79a51
04f5de7a49c66b6cae010d104211e733b362d638
35650 F20101221_AAAWSP miller_m_Page_047.QC.jpg
79a6bfa5e8d0b31b10c41931f374f377
f5a7e327baf123d5c6e2bfdd69c0bd51a4509749
1977 F20101221_AAAVPN miller_m_Page_018.txt
0233a0a13e31981fa7888453d4708ee6
931772f1e6d861b0bd6d8135df4be57563477f42
124000 F20101221_AAAVOZ miller_m_Page_038.jpg
f6d22e8fda0019ebd10f672ecb1cc86f
8e821e96962cc1f62e31601e38b4178e8db8d595
14206 F20101221_AAAVQB miller_m_Page_125.QC.jpg
172990c4d1b391a0a7c897c3500d57c7
f4ed716bd380f8c04cc7b211b49dce0463c89819
34904 F20101221_AAAWTE miller_m_Page_092.QC.jpg
21fcb6126624516b3bbceb6064628306
cdf4a266357df62c1541d822ccd93c6bc50acece
36960 F20101221_AAAWSQ miller_m_Page_019.QC.jpg
8a001ff570de5e61b6416a4c31f5891e
84d1ec18277e509fef2ca64be0f5c814a8a7a568
57418 F20101221_AAAVPO miller_m_Page_016.pro
a8724a0197035fce7f3a94cc1a7359a2
0612005e6dace3de679a4022f35907f2fed063b2
8882 F20101221_AAAVQC miller_m_Page_031thm.jpg
12c85922d66655b181339099b9c349f4
4923b508207102239243d4492b4940bb0f49559f
31861 F20101221_AAAWTF miller_m_Page_068.QC.jpg
1f9b270db7a4481ce6686ec48efc1ca3
1d566a0081ef8a28559fab923ce124de68593500
F20101221_AAAWSR miller_m_Page_015thm.jpg
30eff05bb4a32587e7b0dc6b7a775ec9
3a6eb0436aea33c6a6b3ca1508affb8e31fe66d9
109120 F20101221_AAAVPP miller_m_Page_101.jpg
5a3d5e077a3756f595f439839f8a3cfa
d1ab8fb6b13503a8e6e6229b7a1b155ce3f523b2
507387 F20101221_AAAVQD miller_m_Page_078.jp2
c6d9936c76727c1b7d6d897d9a94c5d7
3a5c39ba4b604c3200f5f9cdad8d5f0aba0d79ed
9081 F20101221_AAAWTG miller_m_Page_044thm.jpg
df8af3d1235920ec683cedd1e0f389e5
7bc37be9c23c8e21b2528d3035390c31dc335712
9082 F20101221_AAAWSS miller_m_Page_089thm.jpg
203e76f0ff0bff65ea692c796a99e0f6
f2d223dc85d37f62def7d7f6d95ed4d1d38127e5
2060 F20101221_AAAVPQ miller_m_Page_107.txt
070eefffabe500eecaec3aa821cb74aa
fb110c9d46e28b9cf8b58ad60970c304b91c9935
63110 F20101221_AAAVQE miller_m_Page_073.pro
c0a8dfc5037d9989ee4f3f19ed2da7e4
b3087bff3ff5e56dc2566eb033f931a9aff06a4f
11629 F20101221_AAAWTH miller_m_Page_126.QC.jpg
18d2bcf15868a1c747978f5c22c0b933
e75b5d29df3109af9dff9f26022c754e0b61b291
36608 F20101221_AAAWST miller_m_Page_091.QC.jpg
e9953b7c967c92db2c5d10fcb991d390
8c8a3a47bc6a15f686024c0107aba3905d58786f
3746 F20101221_AAAVPR miller_m_Page_077thm.jpg
da416a30ea72e7c1ceaadfa552e0b545
f9e2f128679a142e58331ba9859f27cda4b2d65a
F20101221_AAAVQF miller_m_Page_095.tif
647c292c38be999fa66afbf794b8844d
1a9dd05be64501e82f4aaac2f6653b535e95af55
13386 F20101221_AAAWTI miller_m_Page_078.QC.jpg
a7cc473e472034b9836c61c894bef7a2
3e759d461a80c87a5f452f8733edb05eeda515f7
9117 F20101221_AAAWSU miller_m_Page_027thm.jpg
117af4954383e7b0cee17bdc175af95a
a766b8d01621b184a91892ff44a99ede20c80b62
F20101221_AAAVPS miller_m_Page_075.tif
7cfbfece010f42d8e66801673dc42f7c
a5d8072699d51e555f0502dffe9c483b5ea66ec6
2264 F20101221_AAAVQG miller_m_Page_040.txt
1580dcecf5f1de44bae60ca1412ba4f9
a4bf2fe9b91d94b470fb1b4c7893d89a0dd79291
39181 F20101221_AAAWTJ miller_m_Page_123.QC.jpg
517cd8c80e385b930dc0ae94bf889780
daad7bdbf072ef0e2832a56cc6f2d70e55b61729
18582 F20101221_AAAWSV miller_m_Page_007.QC.jpg
5bf7503dad3695a0bbeedbf477af4910
a13d77aa87e2d9c0696425f42915ac9c5004f581
57527 F20101221_AAAVPT miller_m_Page_058.pro
8cb89e7c2c036ed25233d27a893414a6
17cdd1b9dd07af1a904d59b19d87bd73bf99e47b
550 F20101221_AAAVQH miller_m_Page_002thm.jpg
aac694e455c3ba394594e06f2e15b5e3
4f277252ac1050dea2ae71745dbeee808d1ec64b
8849 F20101221_AAAWTK miller_m_Page_038thm.jpg
57fe5cab48d71a192fb56d2fc817e4e6
fb25fa4c71ee82f2ede0be90d295b8adf84bcdd0
33642 F20101221_AAAWSW miller_m_Page_106.QC.jpg
e80d5578ced1c3c7b483c5ad5a608d60
78a9f8513f9e26d4bfc640d1d1ec28ca65af0f7e
F20101221_AAAVPU miller_m_Page_004.tif
397a5131c2c116d10aa51e170a469afd
e88c5fdd55210db1fd7075be96b3978d228f4281
F20101221_AAAVQI miller_m_Page_090.jp2
77d63ed2258cfc7ba01a2f7618d469ed
a9dd8f6ac24beed3187bcdaec00a37ce719b919d
36041 F20101221_AAAWUA miller_m_Page_021.QC.jpg
51fad4dd7adde082fd26c08553b7d9bb
9e725cf4fab9325ad2ad0a9ccc73c6db6a69cf6e
8762 F20101221_AAAWTL miller_m_Page_094thm.jpg
04ba3918075c5553a678045ad1d77442
291ae10254eb9ada0dea7241a83caaa31ba9f9a0
3151 F20101221_AAAWSX miller_m_Page_075thm.jpg
f42181cd9547da756a6b368c8a17d923
65381b1754fdd867be3ccdfefcc2c99f54272a68
F20101221_AAAVPV miller_m_Page_092.jp2
2e747491e2687a3cc91d0d3ddf78e824
790efc8ccefae4499b23cfb56fbe73d09279bb87
8540 F20101221_AAAVQJ miller_m_Page_034thm.jpg
2b600531338286209d4cf1f6e953b9ca
0a76ee1e069980324193a74ca62b1bf4c9c70ea5
9039 F20101221_AAAWUB miller_m_Page_021thm.jpg
09377881601da1053f3e9951f37bb142
13ad4921c1b7d86cf14a42c5991d418016267446
9399 F20101221_AAAWTM miller_m_Page_123thm.jpg
cee004735ebe16bddf924daae7f72614
4665906e45ee70c482128f4197fe109c7777cd6d
37379 F20101221_AAAWSY miller_m_Page_016.QC.jpg
8f8a07f58813f87127d32175f78be971
57fe2ee251db56e58b44207e52a179a847f5658a
8914 F20101221_AAAVPW miller_m_Page_090thm.jpg
95def1f5dcda9b45ce8cccfdc5f9c4f0
ef060dd11dc6b0df61612bdd7c147a65d4258bd5
59295 F20101221_AAAVQK miller_m_Page_079.jpg
b41abf5d5a0be6d0f49b3560bc66f3d3
d9a54f649244d84a2b8007a401eee60bf57597f7
3021 F20101221_AAAWTN miller_m_Page_084thm.jpg
eb84f29c156c4a99a4a28b08927fd1ac
0bd1cb099a508a8bda69cf1f048a316603d046aa
12237 F20101221_AAAWSZ miller_m_Page_119.QC.jpg
f85061820d0eaf4207ff75472d86427f
4749f500269d3a3eff5c5de97fa2e5943475312f
2221 F20101221_AAAVPX miller_m_Page_098.txt
c3b041970f0673ee29f82d30c166a64a
cf415b8fbff2aee7561461d2bf91e1086f529f47
8379 F20101221_AAAVQL miller_m_Page_049thm.jpg
f8e28011cb8b947aad988943d1a8f83c
08b2825d54722df5d9d733020abaeb2cbfe4b5f6
9084 F20101221_AAAWUC miller_m_Page_023thm.jpg
fd913e3f4a3dd268d47837bfffb8a3de
a6d2534c8e60017a9f1bcc9af0fb9d47b8f31ce9
9273 F20101221_AAAWTO miller_m_Page_096thm.jpg
8745193c6580579584e673aaf084f71d
91f44d94889d0a135f50f57757fa829dd23b40a0
F20101221_AAAVRA miller_m_Page_045.tif
9ab23edc4c9577614b009136e9dcb420
a6b6b0bb967c5557e800e82f80d121a41538ffa0
54710 F20101221_AAAVQM miller_m_Page_047.pro
aee04060d703d4a52240544614c67987
60e6b126f495d80f9474ca94b05ef0279e3b1e1b
56361 F20101221_AAAVPY miller_m_Page_103.pro
f168b957868daabc7f6928877a71298d
b14b8a924d090f8d11bc8a28e28abe10ae17f19e
35085 F20101221_AAAWUD miller_m_Page_026.QC.jpg
a3e04c3329a8345588e75f35329d142f
f905fd766bfd97a391ca9ad6ed12e5ac6f981a93
36855 F20101221_AAAWTP miller_m_Page_122.QC.jpg
774b65206e2da8a0eedb4b6ed8c07899
78d280c008df7de2ebf12dafb9306cbbeb7b8c22
8951 F20101221_AAAVRB miller_m_Page_072thm.jpg
6d7dd5de7f9ecc377901085a1372107b
dabcfde688ed3311ecb866012860ca0bd503340e
114539 F20101221_AAAVQN miller_m_Page_024.jpg
98f14f55b40127fb9d62f44b701304ef
b0b8741635df3d961b8404928b9bc84b3d83abc2
86846 F20101221_AAAVPZ miller_m_Page_116.jpg
d9d72b37e39c1e249b656dad66b47d0b
5071f8bf1818bfc6ab5783d213f24b7bd35d6d5c
8588 F20101221_AAAWUE miller_m_Page_026thm.jpg
b61b293ded1f036d3cfc9e6a5ed58e61
df65495fdd6a7d9f54cb29f2024b1522fe833e93
7120 F20101221_AAAWTQ miller_m_Page_118thm.jpg
aecf6e25e05d0ad3f8eb633a140665bc
6fe6d914ff7ccd85dd718d549e5d34b6ffa7f32c
5114 F20101221_AAAVRC miller_m_Page_105.QC.jpg
bdc7d6a34d856b7a0436da496c7ef5d3
5c571741f21d69b753037cec295fb66aea1f08c5
F20101221_AAAVQO miller_m_Page_104.tif
9dbeef137146848335385368bd0ce509
cc2a47a1cee9a01ef5da1daa3ad1634d7a6076fe
8811 F20101221_AAAWUF miller_m_Page_028thm.jpg
268d14b72f4b993e76a691cc3199f66f
1ed3072e6095b07a0f870fd8888dc5c77f634e9b
3961 F20101221_AAAWTR miller_m_Page_121thm.jpg
8a210191c6ab2062dd859474ac2fd07f
44be38b0244def4f058e4b30611cfe4f8839ce91
F20101221_AAAVRD miller_m_Page_041.tif
7e24cc645352dc97dc54bedbbccd8ca2
71d5ab283da4864d03ce973644f6f73dc7ae3ee8
56961 F20101221_AAAVQP miller_m_Page_071.pro
5bf3ad8e993cabfa0202810844acd3e5
1ff6b4ffd0f7b63d12579d53b6d41eeef55edfcb
37113 F20101221_AAAWUG miller_m_Page_033.QC.jpg
555780104ac066adb57f4be92cffaff7
ddd9119bdfb841a2d9aa8edfe6ef81c81afb9c02
8428 F20101221_AAAWTS miller_m_Page_057thm.jpg
ac629960d5fdc0012cdfb3a98f043ccc
1f424acdf8685a9bd190c7e167047aad93631bc6
2207 F20101221_AAAVRE miller_m_Page_030.txt
1ab285b5af62cd83b28434614cfea15e
d63ea68347194eff7bb1277eee969eade4aeb9f5
2615 F20101221_AAAVQQ miller_m_Page_122.txt
a9493cb8543d6b3e44054c1a557a6b8a
c34a671d5758fb8d26e23b5a30a6ea5c1065cb36
13337 F20101221_AAAWUH miller_m_Page_041.QC.jpg
47db80801fadd8e67f26bb75cf2f1c5b
4908f9c811e900cdc9635f6b8600701827bd9619
16020 F20101221_AAAWTT miller_m_Page_079.QC.jpg
a2cc7739a5fbd67f337aa21a96801e9f
eaec315c0a309712bf4fc08817d72c941761a3dc
37550 F20101221_AAAVRF miller_m_Page_071.QC.jpg
9c16991a86c42f9428a2155afa44e655
b1114c6dfc7bbe5562ca7769b8f63e535719d962
111978 F20101221_AAAVQR miller_m_Page_062.jpg
d96972527493bb5cc7daadd7eb5821a6
88b79acadf694b1dcb737877131928bb9658a4ea
9160 F20101221_AAAWUI miller_m_Page_046thm.jpg
7d7fff98c1805699a28cb8935317c994
7bf031e0748e7530a0f7338e45391daf104eec05
189153 F20101221_AAAWTU UFE0022335_00001.xml
622f8df5320c6ddb40ae792950a149ef
ce94294920f60fa8bd9e398f95f3e94179c2f2f3
36285 F20101221_AAAVRG miller_m_Page_060.QC.jpg
082d44c520712cda1b5feef5170317c2
0418cf6bff33c985f0b1dd8f3f64e066dea6137f
F20101221_AAAVQS miller_m_Page_102.jpg
e2fcd68e51462b102bc66500e4efcf5b
ae94b868f5aef112465771e2dc9d0ac46c6d55f1
37944 F20101221_AAAWUJ miller_m_Page_050.QC.jpg
9e6ad5597bf9349e463222a558ecc373
c2e78d829ad6a7727986f88d9ce90851a783fabc
6133 F20101221_AAAWTV miller_m_Page_005thm.jpg
e023586933b98feecf4113782fc9b109
4a3d578dc5c546f63facfbb2904f9dabc1d23678
4772 F20101221_AAAVRH miller_m_Page_109thm.jpg
1a5ae018b1e4ee5124258fa2fee28411
01b4e8f7050c274fb6f8836696944f99d2388e50
55563 F20101221_AAAVQT miller_m_Page_065.pro
b3e228660ec69df8802407c4859ecc9c
9738941ba7611769d9f22809fbd6ba05494a093b
8837 F20101221_AAAWUK miller_m_Page_053thm.jpg
0c837f6746271da10621cc71c48cd390
a683962f2a21b1dee00fa5dd7958d13ae9a7c0cd
9318 F20101221_AAAWTW miller_m_Page_008.QC.jpg
98d44e3d8e25c029c6c9151da8befc5e
5b43d1b0bf8602ee8ecc9bd4b9af5f32e8017d8b
2151 F20101221_AAAVRI miller_m_Page_003.txt
ad369b121bed89bef650c80f6c1b1dc5
97b69b241bcceb3c6be0eba72e52f0f4a8f218a5
34892 F20101221_AAAVQU miller_m_Page_100.QC.jpg
2927e2123f53eda29ec6885b6dcc9c30
eec50ece97c6865d52200bf882212a2988c30b84
37800 F20101221_AAAWUL miller_m_Page_063.QC.jpg
0d477bb181580e31bfc2befc5a8f7423
6570e71c77a403faec3035042f01bc917a01cc43
31983 F20101221_AAAWTX miller_m_Page_009.QC.jpg
160502bacd34121ef43287256b5ea5a9
5378f38df5598a25bfcf857292275840af595a8f
123864 F20101221_AAAVRJ miller_m_Page_105.jp2
45c8238139c98cdf11e510635b0aabef
b9e987926d01b1a23dd1a666d734a001e78f44d4
F20101221_AAAVQV miller_m_Page_010.tif
f2777fa3bd2a7f0a2e0aa959f8c83bc1
4df4452f7824e7ab5874956e2b9919bae3de1cca
38774 F20101221_AAAWUM miller_m_Page_073.QC.jpg
a323d6c00f7ee0e855ff790876e8685d
ca3c9fe552f58287ec4e823eb1d2070c0ff82a52
31091 F20101221_AAAWTY miller_m_Page_011.QC.jpg
07e324732b56c394763a16f41ae19612
6b880b20b0b888dcbeea6aaa3ff7dd4facc0a9b8
F20101221_AAAVRK miller_m_Page_032.jp2
35337017d2498669ace213874345fceb
e5f221f45990e5395fb2dc5a055d022c66c6d01c
37360 F20101221_AAAVQW miller_m_Page_031.QC.jpg
f20229ddc694908e94be3a0bf79da393
493fcf890a0bfdb39c0f210f719ca1a9c7020cd2
3496 F20101221_AAAWUN miller_m_Page_078thm.jpg
5edd6d3066bc26cdffdd313ef4b573be
204d385b3f37eaa2119243231fcc7055f7bb83bd
9050 F20101221_AAAWTZ miller_m_Page_013thm.jpg
ca5c8b2f18a57de8fabd99fb71c738e9
63f289b37036742c0478c0262c355f157ff1a063
55715 F20101221_AAAVRL miller_m_Page_098.pro
2c0335e90b25416f3a606302dc68b033
20c1cd65fbdddde498795de75320701d8d9c9b30
F20101221_AAAVQX miller_m_Page_098.tif
5d98df855b32f0d940fa7313a037eaf6
18ef3338732df7b727b88bb3f72f4a778feb29be
49923 F20101221_AAAVSA miller_m_Page_018.pro
01e6f7053fbb1f83a76e2b0376f75f46
81b3b9c83339fb2267ff74ff689acd9caf04d5ee
13935 F20101221_AAAWUO miller_m_Page_080.QC.jpg
2bace8b9fb7112c337e80cc2c799e4dc
e8d94cb80d0a8ba9babaae5fbb605f44baf4d8d2
36948 F20101221_AAAVRM miller_m_Page_012.QC.jpg
73414e35f7dcae6a0fc28d53c3875e12
4b974ece3e63254fba8a74a9c951c216008eb145
396321 F20101221_AAAVQY miller_m_Page_084.jp2
e9639f2ba3ad2629fa6335d7cc7071d1
30c5c2fa47c60d30e3a7ee54e033602290acbfe5
3437 F20101221_AAAWUP miller_m_Page_080thm.jpg
cecb3a756202cf10bb9a38fdd3311d23
4fb742240c46617b373a3b4cac4fe3c6a0af00a5
107524 F20101221_AAAVRN miller_m_Page_052.jpg
2247f9c9692391997dad09b04a86136f
645caf5050ca70f4d356b6569394d8539515353b
32637 F20101221_AAAVQZ miller_m_Page_018.QC.jpg
e59b9b64b3987f90560f236a5fd1584c
e29378c78532b5175c14a23655ff90c54edf450a
109438 F20101221_AAAVSB miller_m_Page_029.jpg
8e7e7bf045a7f8b44723d7538738b279
34633b934b7fd1d6b5c38d7b342490ac35950edd
8551 F20101221_AAAWUQ miller_m_Page_095thm.jpg
f5c82ecb47c189d04e8f7b9c443a89af
0212e690a942deaa2abfff3274ae94f30e61ca98
F20101221_AAAVRO miller_m_Page_016.jp2
0dad66d6de4b04392ec15494cb377e9a
58572cf19912ef513b3a183922867f5f63b2989e
36375 F20101221_AAAVSC miller_m_Page_072.QC.jpg
be41f8ab2acd64cb86e78fc0d6119070
4fb5250000f83807def42e9eae46b88a825a81ad
37784 F20101221_AAAWUR miller_m_Page_098.QC.jpg
2a01584987040b27f081ac885e8bf391
dc026e3bc8b76972a218f9bdc0cd65516c3b06dd
2719 F20101221_AAAVRP miller_m_Page_111thm.jpg
974fd347863fa8fd4f699b89d8f068bb
a94bdb8cf48aac75c638f29b39dd36ff242b0e99
853 F20101221_AAAVSD miller_m_Page_084.txt
8be315ec98f30f296a9f642820a668e3
bc151085d89f860d0a9ab7d78116464a2502ed97
F20101221_AAAWUS miller_m_Page_098thm.jpg
c81bc71bd26bdcfa7c192ba5ae0834e8
03f0b1584c24fe105b67a2e48ac8215d0bc8ea79
9542 F20101221_AAAVRQ miller_m_Page_085.QC.jpg
a32110c17ce1d88e5bb7ebfa9aa68a9a
85bedf00a76161b5080c625f453effe40fa57592
14733 F20101221_AAAVSE miller_m_Page_077.QC.jpg
008537866aa8732e1be9e5da912f9558
485b6d1a51cf2920b4bdc4cda471115bd5dfe0a1
37461 F20101221_AAAWUT miller_m_Page_103.QC.jpg
3e58b73b71ad195f4da5633aeecf67b7
7acc3784ed9ee5f7e418fdbec5601c7a299424b8
54208 F20101221_AAAVRR miller_m_Page_101.pro
91e2b65cc6b4925e1c4545b470f1652d
e94ca43cfd8f4c65155fadc7e374371cb5dff388
35039 F20101221_AAAVSF miller_m_Page_039.QC.jpg
7affef184f1e61e30cd1c10b5ca890bb
4201ebf82a34432849725082e1e5680ad49d2c69







COVERT AGGRESSION:
THE MEANS AND MOTIVE OF "GETTING AWAY WITH IT"

















By

MARISA L. MILLER


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

UNIVERSITY OF FLORIDA

2008






































O 2008 Marisa L. Miller









ACKNOWLEDGMENTS

I am immensely thankful to the many people who supported me throughout this process.

First, I would like to thank my supervisory committee chair, Dr. Barry Schlenker. From the

theoretical seeds of this proj ect, he helped nourish these ideas and my excitement for them. With

endless patience and vast wisdom, he helped me make this proj ect what it is. In this proj ect and

throughout my graduate training, he broadened my knowledge and deepened my understanding

of scientific psychological research, and for that I am immensely grateful. I would also like to

thank my other supervisory committee members. I thank Dr. Ken Rice for his point of view that

helped broaden my thinking. I thank Dr. James Algina for his endless willingness to answer

statistical questions, and to help me navigate the often frustrating world of SAS. I would also like

to especially thank Dr. James Shepperd for his sense of humor, his encouragement of these ideas

in their initial stages, and his honest enthusiasm for this proj ect.

I would also like to thank my friends and family who gave me the support I needed. To my

parents, Kathy Davis and Bill Miller, I thank them for instilling in me an interest in thinking

about the world, and a fascination with understanding and thinking critically about the behavior

of people in this world. To my graduate school friends, Jodi Grace, Penny McNatt Devine,

Melinda Goodman, Meredith Terry, Kate Sweeny, Wendi Malone, Gizem Aksoy Toska, Ryan

Johnson, and others, I thank them more than they know for their immense support,

commiseration, humor, empathy, and friendship--always showing perfect understanding,

sometimes without knowing that it was needed. I would also like to thank my longtime friends

and extended family, Kristen Miller, Will Miller, Barry Davis, Katy Nichols, Erin Zier, Venussa

Kassees, Maryliz Pinto, Nessa Manten and others, who gave me a fresh dose of reality and kept

me grounded in the real world outside of graduate school, and showed unrestrained and constant









faith in my ability to succeed. I particularly thank Katy Nichols for enduring with me the

experiences that inspired this proj ect.

Finally, I would like to thank coffee. Despite my initial distaste for coffee 5 years ago, I

grew to crave its caffeinated goodness. I would not have survived graduate school or this

dissertation process without it.












TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............3.....


LIST OF TABLES ................. ...............7..____ .....


LIST OF FIGURES .............. ...............8.....


AB S TRAC T ......_ ................. ............_........9


CHAPTER


1 INTRODUCTION ................. ...............11.......... ......


Aggression Taxonomy ................. ...............12.......... ......
Physical versus Verbal .............. ...............12....
Active versus Passive ................. ...............13................
Direct versus Indirect .................... .. ............ . .......1

Special Note on Terminology: Indirect, Relational, and Social Aggression. ..................1 7
Gender Differences ................. ...............19.................
Intelligence and Indirect Aggression ............... .. ...... ........ ...............20. ....
Self-Presentation and Reputation in Aggression's Aftermath ................. .......................24
Situational Factors .........._..._._ ...............3.. 1.............

Hypotheses............... ..............3
Current Study ................. ...............40.......... ......


2 METHOD .............. ...............42....


Overview ................. ...............42.................

Design ................. ...............42.................
Participants .............. ...............43....
Pre screening Session .............. ...............43....


3 RE SULT S .............. ...............49....


Pilot Testing ................. ...............49......__ ......
Feedback ..........._...__........ ...............49.....
Message Options ............... ... ...............50..
Scoring the First Impressions Questionnaire ........_........_...__ .........._ ...........5
Overview of Analyses. ..........._..._ ...............53..._..__......
Messages Used...................... .... ...............5
Positive, Negative and Alternate Messages............... ...............54
Category Comparison ..........._...__........ ...............56.....
Summary of Message Use ..........._..... ... ........_._ ............._ ............6
Obj ective Assessment of Design and Evaluation Quality ................. ............ .........63
Interaction Expectations .............. ...............64....












Perceptions of Partner............... ...............6
A effect ............... ... ............ ...............69.......
Communication Preference .............. ...............70....
M oderators ................. ...............70.................
Perspective Taking .............. ...............71....
Other M oderators ................. ...............72.................


4 DI SCUS SSION ................. ...............87................


Comparisons between Positive and Ambiguous Feedback .............. ...............89....
Nature of the Interpersonal Relationship ................. ...............94........... ...
G ender. ................... ... .......... ......... .............9
Expectation of a Future Interaction .............. ...............98....
Social Desirability .............. ...............101....
Perspective Taking............... ...... ...................10
Distinguishing Covert Aggression from Other Forms of Aggression ................. ...............103


APPENDIX


A PERSONALITY PACKET (PART 1)............... ...............106..

B PERSONALITY PACKET (PART 2). ............ ......_ ...............110.


C BASIC INFORMATION QUESTIONNAIRE ...........__ ...... __ ......._..........11

D FEEDBACK ............ ............ ...............112...


E FIRST IMPRESSIONS QUESTIONNAIRE (MIXED MESSAGES CONDITION) .........113

F FIRST IMPRESSIONS QUESTIONNAIRE (NEUTRAL MESSAGES CONDITION)....114

G PRELIMINARY INTERACTION AS SES SMENT ................. ................. ......... 115


H STATEMENT SELECTION MEANS ................. ...............120...............


I STATEMENT SELECTION MEANS WITH OFFSET COMPARISON GROUPS
NOT EXPECTING A FUTURE INTERACTION ...._.. ................. ........._...__.....121


LIST OF REFERENCES ........._.._ ......___ ...............122....


BIOGRAPHICAL SKETCH ........._.._ ..... ._._ ...............126....











LIST OF TABLES

Table page

3-1. Pilot study: Perceptions of feedback ................. ...............75........... ..

3-2. Ranking means ................. ...............76........... ....

3-3. Category ranking difference scores .............. ...............77....

3-4. Ranking means with offset comparison groups not expecting a future interaction ...........78

3-5. Trait groupings and relevant Cronbach alphas .............. ...............79....

3-6. Perceptions of partner and affect .....___................ ............... ........ ...._...80

3-7. Participant communication method preferences ................. ...............81........... ...

3-8. Ranking of negative messages as a function of perspective taking, feedback, and
response option .............. ...............82....

3-9. Ranking of positive and negative messages as a function of social desirability,
feedback, and response option ................ ...............83................

H-1. Statement selection means ................ ...............120...............

iI-. Statement selection means with offset comparison groups not expecting a future
interaction .............. ...............121....










LIST OF FIGURES


FiMr page

3-1. Mean ranking for negative messages as a function of perspective taking, feedback,
and response option............... ...............84.

3-2. Mean ranking for negative messages as a function of social desirability, feedback,
and response option............... ...............85.

3-3. Mean ranking for positive messages as a function of social desirability, feedback,
and response option............... ...............86.









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

COVERT AGGRESSION:
THE MEANS AND MOTIVE OF "GETTING AWAY WITH IT"

By

Marisa L. Miller

August 2008

Chair: Barry R. Schlenker
Major: Psychology

Existing aggression literature covers a variety of aggressive means and motives, but so far

little research has examined the nature of covert aggression-aggression that simultaneously

contains both a positive and negative interpretation, thereby possibly escaping the negative

consequences of such aggression. This study stands as the first experimental investigation of

responses to covert aggression and the use of covert aggression. Participants (N = 295)

completed a bogus creative intelligence assessment and received false feedback from a partner

(actually nonexistent). The evaluative feedback was either positive, negative, or conveyed an

ambiguous (mixed) message. Participants then had the opportunity to send a combination of

positive, negative, and alternate messages to the evaluator. The alternate messages were either

ambiguous, and thereby provided an opportunity to employ covert aggression, or neutral, and

thereby provided no opportunity for covert aggression. Participants subsequently reported their

perceptions of their partners and emotional state. Pilot participants verified that the feedback in

the ambiguous, positive and negative feedback were perceived as containing mixed, positive, and

negative interpretations, respectively. In general, participants did not respond negatively to the

ambiguous feedback and viewed ambiguous and positive feedback partners in a similar, positive

manner. In contrast, participants receiving negative feedback were subtly affected by the









availability of mixed messages, indicating that when people can send mixed messages, they do

so and are, on the surface at least, nicer than when they must express hostility in an open and

clear manner. Negative feedback participants experienced more negative affect and reported less

favorable views of their partners than either positive or ambiguous feedback participants.

Individual difference measures moderating these effects are discussed.









CHAPTER 1
INTTRODUCTION

If you can't say anything nice, then at least have the decency to be vague.

-Unknown

History books chronicle the existence of humankind' s aggression pitting one civilization

against another with history-changing consequences. Without a doubt, wars, shots fired, and men

killed have certainly had long-lasting impact on societies throughout the world. However, just as

many epic outcomes have resulted from far less conspicuous conflicts. Governments, militaries,

and political groups have used spies throughout history to garner information, complete

assassinations, and to aggress against other governments, militaries, and political groups in a

covert manner that allows the sponsoring agent to preserve an image of dutifully abiding by

formal agreements. Government spies frequently adopt and portray alternate identities that gain

the confidence of others, gain access to important information, and conceal the spies' ulterior

motives.

Although perhaps less likely to be recorded in history books, this subversive and covert

form of approaching hostility and aggression that maintains a surface appearance devoid of

malicious actions is just as likely to have important consequences for political groups as more

conspicuous and physically violent actions. In a similar manner, individuals in everyday life also

desire the appearance of civility while secretly devising and enacting hostile activities.

Maintaining the appearance of one thing while doing another is a useful but potentially

dangerous path to choose in interpersonal relationships. Unfortunately, little to no research has

directly assessed interpersonal aggression considering this covert motive. Instead, much

aggression research has ignored identity concemns-how the would-be aggressor appears to









others and manages his/her identity in the aftermath of aggression. In short, he/she must actively

disguise his/her aggressive actions by engaging in covert aggression.

Although aggression comes in many forms and guises, social psychologists define

aggression as behavior that is intended to cause harm. Previous research highlights several

dimensions by which aggression can be categorized and also pinpoints specific personal

characteristics and situational determinants of aggression. While some research delves into

territory and behaviors that may be dually characteristic of the covert and overt aggressor, a

covert aggressor faces challenges distinct from those of the overt aggressor, and there is reason

to conclude that unique processes are at work as the covert aggressor acts in his/her social world.

To fully understand the unique struggles and considerations of the covert aggressor, it is

necessary to first have a proper understanding of the more prevailing aggression research before

focusing on indirect aggression, a similar but still distinctly different type of aggression.

Aggression Taxonomy

A pioneer in aggression research, Buss (1961) developed a taxonomy to categorize various

types of aggression that served as a guiding structure for subsequent generations of aggression

researchers. When considering aggression, the most salient examples include beating, shooting,

shoving, stabbing, etc. In fact, the majority of psychological studies have assessed physical

forms of aggression by such measures as willingness to shock other participants or blast them

with annoying and painfully loud noise. However, aggression also includes behavior such as

verbal assault, rumor spreading, and ostracism. To capture the vast array of aggressive actions,

Buss proposed three dimensions: physical vs. verbal, direct vs. indirect, and active vs. passive.

Physical versus Verbal

Although physical aggression may be the most obvious example, Buss (1961) emphasized

that physical acts are not the sole means by which individuals can be aggressive. Verbal actions










can harm as well, but verbal aggression causes psychological harm rather than physical pain and

injury. Specifically, Buss proposes two types of psychological pain: threat of future violence or

rejection, with rej section incorporating any type of response that labels the victim as aversive,

bad, unwanted, or lacking in some way. Some common examples of verbal aggression as

rejection include aversive teasing of an unpopular peer, publicly denouncing a person's

character, fabricating lies about another person, derogating another person in an argument, etc.

Active versus Passive

Buss's (1961) original taxonomy also included the difference between active and passive

aggression. Buss noted in his brief mention of this dimension that aggression can vary in the

degree to which it is actively vs. passively committed. Specifically, active aggression requires

taking initiative to construct an impediment that prevents the victim from satisfying a goal,

taking action to hurt another, etc.; in other words, it requires some degree of effort on the part of

the aggressor to deliver noxious stimuli to the victim. In contrast, passive aggression does not

require the aggressor to act and in fact, the very fact that the aggressor is not acting may

constitute or allow the aggression. In other words, the simple presence of the aggressor or the

aggressor' s inaction may be aggressive. For example, failing to render help may be seen as

passive aggression. Witnesses to a crime who fail to aid a victim may be accused of passive

aggression if the witnesses intend this to be the side-effect of their inaction. Using this definition,

it may be possible to classify some witnesses to the famous Kitty Genovese murder as passively

aggressive, especially as her homosexuality may have given her neighbors sufficient aggressive

motivation. Consider another example--a passive aggressor may fail to move out of another' s

path simply to irritate the victim. Passive aggression is likely to be a successful strategy for the

person motivated to avoid repercussions. Buss (1961) argues that "Passive aggression is a

subordinate's best weapon against his superior. Active attack invites retaliation. When the attack










is passive, however, it is usually difficult for the victim to establish blame or to determine

whether aggression has occurred" (p. 9). Unfortunately, social psychologists focused primarily

on active aggression with little research attention to the more passive and subtle forms of

aggression.

Despite social psychology's neglect of passive aggression, research on passive aggression

is quite popular among clinical and counseling psychologists. Research on passive aggression led

to the classification of the passive aggressive personality disorder (also known as the

Negativistic Personality Type). According to this conceptualization, passive aggression is seen as

a means of aggression for individuals restricted in their ability or willingness to confront another

in a direct and open manner. The Millon Clinical Multiaxial Inventory-III (MCMI-III) described

the passive aggressive personality type as an "ambivalent orientation" in which people "are torn

between following the rewards offered by others and those they themselves desire....These

individuals experience endless wrangles and disappointments as they vacillate between deference

and defiance, obedience and aggressive opposition. Their behavior is characterized by an erratic

pattern of explosive anger or stubbornness intermingled with periods of guilt and shame"

(Millon, Davis, & Millon, 1997, p. 18). Some of these behaviors match Buss's (1961) original

conception of passive aggression but not all. Specifically, Buss suggests that individuals with a

need to be covert in their aggression may choose passive aggression, and this is somewhat

reflected in the ambivalent orientation of wavering between deference and defiance, but the

anger and stubbornness are unique to this conceptualization. The primary means of passive

aggression (failing to act rather than actively initiating a disruptive behavior) clearly establishes

1An abundance of research on the topic led researchers to abandon it as a useful
classification because criteria for diagnosing the disorder were imprecise and hard to
sub stanti ate.









difficulty in assigning blame for harm to the victim; in determining cause for some event, people

look for changes or actions deliberately taken. Failure to act is a more difficult basis upon which

to cast blame. For example, failing to help someone could be a form of passive aggression and

more difficult to pass blame for such a behavior versus casting blame for physically assault.

The MCMI-III (Millon, Davis, & Millon, 1997) also characterizes passive aggressive

people as expressively resentfud by resisting the fulfillment of others' expectations,

procrastinating, etc., and other behavior that receives gratification in the demoralizing and

undermining of others' pleasures and aspirations. These individuals are also thought to have a

discontented self-image in that they see themselves as misunderstood, luckless,

underappreciated, and demeaned by others. The implication from this depiction of the passive

aggressive personality disorder is one of a person who feels deserving of the appreciation of

others but does not believe that he/she receives it to the degree deserved. In a sense, then,

although they realize that they are devalued by others, they feel this devaluation is undeserved as

juxtaposed with their own high-value self-assessment. Therefore, the passive aggressive person

may be somewhat narcissistic.

The characterization of the passive-aggressive individual as someone who believes that

he/she is underappreciated and demeaned and may resort to behaviors that thwart others implies

two things. First, the characterization of someone who is demeaned implies that the passive

aggressive individual is in a role of equal or lower status to those demeaning him/her. Moreover,

the ability of this person to thwart the goals of others implies that he/she plays an active role in

the lives of others. In short, individuals may be more likely to use passive aggressive when

aggressing against those higher in status and those with whom one expects ongoing interaction.

In contrast, aggressing against lower status individuals poses a lower need for covert aggression,










and so more openly and overtly hostile acts may occur to those lower in relative status than those

higher in relative status.

Direct versus Indirect

The direct vs. indirect dimension of Buss' s (1961) taxonomy describes whether or not the

victim is present at the moment of aggression when harm is inflicted. Direct aggression is that

which requires a face-to-face confrontation, i.e., aggression done in the presence of the victim.

In contrast, indirect aggression is done in the victim's absence. Frequently, direct aggression is

misconstrued as physical aggression and indirect aggression as verbal aggression. Although the

overlap may frequently be true (hand-to-hand combat is considerably easier when in the victim's

presence), it is not always the case. For example, an aggressor may engage in indirect physical

aggressive by adding poison to food that the victim will eat hours later. Similarly, verbal

aggression can be both direct (such as teasing someone) or indirect (by spreading gossip).

Several authors have noted that indirect aggression is a particularly shrewd method to pursue

because it helps to prevent a counterattack: because the aggressor is not present at the moment of

aggression, it is difficult to identify the aggressor thereby hindering retaliation (Buss, 1961;

Bjiirkqvist, Ojsterman, & Lagerspetz, 1994; Bjiirkqvist, Ojsterman, Lagerspetz, Landau, Caprara,

& Fraczek, 2001). In fact, this "behind-the-back" nature of indirect aggression was originally

construed by researchers to mean that the perpetrator would remain anonymous to the victim;

however, in the strictest definition, indirect aggression does not necessarily protect identity. This

focus on anonymity has since been expanded to include the fact that indirect aggression may not

hide the identity of the aggressor, but simply the aggressive motivation for a particular action

(Archer & Coyne, 2005). Bjiirkqvist et al. (2001, p. 113) argue that indirect aggression "attempts

to cause psychological, in rare cases even physical, harm to the target person by social

manipulation, often [but not always] attacking the target in circuitous ways through a third










person to conceal the aggressive intent, or otherwise pretending the attack was not aggressive at

all."

Special Note on Terminology: Indirect, Relational, and Social Aggression

Covert aggression, by necessity, must leave little clear evidence that implicates the

offender. This will likely lead individuals to employ means of aggressing that leave little

physical evidence, are subj ect to ambiguity and lend themselves easily to alternative

interpretations. Of the domains available to the aggressor, physical, direct, and active aggression

are particularly likely to convey a clear intention of aggression. Individuals may choose verbal,

indirect, or passive aggression to aggress because tone-of-voice can be reinterpreted, eye-witness

accounts are missing, and volitional actions are missing. As a result, the domain of behaviors

likely to be chosen by the covert aggressor include a set of behaviors (e.g., aggression committed

through social networks with social implications for the victim--teasing, ostracism, gossip,

sarcasm, etc.) that have been classified in multiple ways. Indirect aggression, in the purest sense

of Buss' s original work (1961), does not imply intention to do social harm or the use of others to

accomplish one' s misdeeds. However, much of the previous work on indirect aggression has

come to be nearly synonymous with aggression of a social nature. Archer and Coyne (2005)

identified three distinct terms that share similar characteristics: indirect aggression, relational

aggression, and social aggression.

The term relational aggress.v~ion was first introduced by Crick and Grotpeter (1995) in their

study of aggression amongst pre-adolescent children. The authors later defined relational

aggression as "behaviors that harm others through damage (or the threat of damage) to

relationships or feelings of acceptance, friendship or group inclusion" (Crick et al., 1999). In its

purest sense, then, relational aggression can be either indirect (such as gossip) or direct (such as a

scorned lover who confesses all the torrid details to the ex-lover' s spouse); hostile/emotional or










instrumental (damaging a victim' s relationship or social standing in a group to create new social

opportunities for the aggressor); or verbal.

The term social aggress.v~ion was originally coined by Cairns et al. (1989) and defined as

"the manipulation of group acceptance through alienation, ostracism, or character defamation"

(p. 323) and later expanded as "...directed toward damaging another' s self-esteem, social status,

or both, and may take such direct forms as verbal rej section, negative facial expression or body

movement, or more indirect forms such as slanderous rumors or social exclusion" (Galen &

Underwood, 1997, p. 589). In contrast to relational aggression, social aggression focuses not on

the means of aggression but rather on the intended harm. In other words, social relations are the

goal, not the tool, of aggression.

Archer and Coyne (2005) argue that these three terms are fairly synonymous and equally

viable alternatives to physical aggression and distinctions between the three terms causes

unnecessary conceptual complications. Although the behaviors that characterize these three types

of behaviors are very similar--if not identical--ignoring the differing motives behind the various

types of aggression, as Archer and Coyne implicitly suggest, can have negative repercussions in

trying to gain a thorough understanding of these behaviors. Behaviors may seem similar at a

surface level, but the reasons behind these behaviors can vary dramatically. However, more

importantly, failing to consider self-presentational concerns and the aggressor' s desire to

embrace both a nonaggressive identity and the efficacy of aggression neglects much of what the

would-be aggressor must consider. Without properly understanding all of the considerations in

the mind of an aggressor, a full understanding of aggression in an everyday context cannot be

attained.









Gender Differences

In early aggression research, researchers concluded that men were more aggressive than

women (Baron & Richardson, 1994). However, upon closer examination, researchers have

tended to since conclude that when gender differences occur, men and women tend to choose

different aggressive strategies that suit their particular gender constraints such that males choose

direct physical aggression and females choose indirect verbal aggression (Bjoirkqvist, Lagerspetz,

Kaukiainen, 1992; Crick & Grotpeter, 1995; Lagerspetz, Bjoirkqvist, & Peltonen, 1988).

Although there is a tendency for men and women to differentially employ certain aggressive

strategies, the reason for this is subj ect to interpretation. One path of reasoning may point to

biological differences between the sexes that lead women to pursue a course of aggression more

suited to their weaker physical strength whereas men have the physical prowess necessary to

invoke direct physical aggression.

However, another path of reasoning suggests that women employ indirect aggression more

than direct aggression because of differential social resources. Specifically, women are more

socially connected than men (Green, Richardson, & Lago, 1996) and therefore face greater

repercussions for direct aggression; e.g., attacking a single individual in a densely connected

social network brings with it a greater likelihood that the victim has social connections that will

come to the defense of the victim (versus attacking a single individual in a less dense social

network). At the same time that a dense social network inhibits the likelihood of direct physical

attacks, it simultaneously provides the resources necessary to utilize indirect (or social

aggression) because this frequently requires the complicity of others (Green, Richardson, &

Lago, 1996). Because men are less socially connected, they have fewer consequences for

utilizing direct aggression and fewer resources to utilize indirect aggression. Green et al. (1996)

propose that the gender differences in aggression type are merely an adaptation made to









contemporary challenges of men and women--when controlling for social-connectedness, men

with denser social networks reported more indirect than direct aggression and less direct

aggression than men with low-density social networks, thereby demonstrating that the key

difference is social connectedness, not gender. Indeed, as children progress into early adulthood,

these developmental differences should decline; this is in fact the case for 23 year-olds

(Bjoirkqvist, Kaukiainen, Laurila, & Ylinen, 1994; as cited in Green, Richardson, & Lago, 1996).

Intelligence and Indirect Aggression

Just as gender-typed social networks can impact the type of aggression one chooses, other

individual differences of a more dispositional nature also influence choice of aggressive tactics.

Several theories contend that one key individual difference is intelligence. For example,

Tedeschi and Felson (1994) argued that individuals with greater intellectual capacity are less

likely to resort to aggression of any sort (versus other tactics such as negotiation), but should also

be less likely to resort to the more brutal forms such as physical aggression. In other words,

strong intellectual abilities allow individuals to circumvent direct physical attacks by means of

more refined verbal or indirect attacks on another. This view elicits the possibility that less

intelligent individuals should show more overt aggression than their more intelligent and

covertly aggressive peers.

Similarly, to predict maladaptive social behavior and aggression, Crick and Dodge (1994)

proposed an information processing model. This model outlines a series of steps that individuals

(in particular, children) take when reacting to others. The social skills deficit model proposes a

direct link between bullying behavior and a dearth of social skills. In other words, individuals

who lack the necessary social skills to resolve conflict peacefully may resort to aggression as

their only recourse.









The model outlines six distinct steps individuals take when acting in the social

environment and these researchers argue that aggressive maladjusted children exhibit specific

differences from nonaggressive socially adjusted children. First, individuals must encode cues

from their environment (i.e., notice what is happening). Second, individuals must interpret these

cues and give them meaning--was the shove from a classmate accidental or intentional? Third,

individuals must select a goal--a desired end state, such as obtaining a toy, making a friend,

getting even, or staying out of trouble. Fourth, individuals generate a host of possible behavioral

responses that may aid the attainment of the goal--being nice to the other person, sharing a toy

with the other person, reporting a transgression, etc. Fifth, individuals must select one of these

generated responses and sixth, enact that response.

In an extensive review of this model, the authors report findings indicating that aggressive

children are more attentive to aggressive cues in the environment, more likely to interpret cues as

aggressive, generate fewer response options, and are more likely to select an aggressive response

option. The seeming resolution to this problem is to teach these children a greater range of

response options, how to better select options with a positive outcome (e.g., teaching children

that playing nicely with others is more likely to win friends than punching their peers), and how

to avoid misinterpreting environmental cues as aggressively motivated. In line with this

reasoning, Bjoirkqvist and colleagues (2001) describe anti-bullying programs that teach children

a greater appreciation of their peers' emotional responses to aggression as a means to better

predict the outcome of their aggressive actions (presumably, rej section by one's peers).

Although this model seems quite logical and matches the popular conceptualization of the

slow-witted bully, it makes some critical oversights in understanding aggressive children.

Specifically, Crick and Dodge (1994) investigated physical and direct forms of aggression by










operationalizing aggression as "the extent to which children start Eights, hit, push, or threaten

peers" (p. 82). Moreover, this social information processing model is an attempt to explain

maladjusted behaviors broadly and the authors perceive aggression as so closely linked to social

rej section that they frequently use the terms aggressive and maladjusted interchangeably.

However, this specific operational definition of aggression limits the model to specific types of

aggression: direct and physical. By failing to incorporate additional behaviors, this model does

not capture the full array of aggressive behaviors because it overlooks the indirect and covert

forms of aggression. These types of behaviors may not fit within the framework Crick and Doge

promote of the socially inept aggressor.

In keeping with this possibility, this model fails to account for data that Eind positive social

outcomes for aggressive children (e.g., LaFontana, & Cillessen, 2002; Salmivalli, Kaukiainen, &

Lagerspetz, 2000). Specifically, these authors found that the use of indirect aggression as

reported by peers was positively related to peer acceptance (Salmivalli, et al., 2000) and that

relational aggressors may even attain higher sociometric status than their less aggressive peers

(LaFontana, & Cillessen, 2002). In other words, an aggressive response option seems to be

associated with at least some positive outcomes and, depending on one's goals, may actually be a

fairly adaptable response choice. These resultant positive outcomes may then increase the

likelihood of using aggression, just as any other reward increases behavior. Therefore, it seems

that covert aggression requires a certain degree of social savoir faire; social illiteracy would

sabotage attempts at indirect aggression. Indeed, other researchers such as Kaukiainen et al.

(1999) even found that higher levels of empathy (the aspect of empathy that involves

understanding the emotional perspectives of others) actually lead to higher levels of indirect










aggression. This tendency may cause the anti-bullying programs described above to backfire by

giving the bullies new tools with which to cause harm.

This seeming disparity between the social information processing model (Crick & Dodge,

1994) and these other authors' work can be understood in terms of the specific aggressive types

the authors are each considering. Indirect aggressors may or may not differ from the direct

aggressors on which Crick and Dodge focused in terms of cue encoding, cue interpretation, or

goal selection, but individuals choosing to aggress indirectly seemingly possess greater insight

into the psychological state of their victims, making their aggression all that more powerful and

harmful. Moreover, this greater insight likely also helps the children avoid the same

repercussions direct aggression would elicit because as indirect, covert aggressors armed with

increased understanding of their victims, they may know better whether their victims will

retaliate, confront, or report them. Finally, indirect, covert aggressors may also differ in the

generation of response alternatives and selection of responses. Bjoirkqvist et al (1992) argue that

aggressive strategies change as children develop new skills; children first use physical aggression

when they develop self-motor control, then verbal aggression when they master sentence

construction and vocabulary, and then finally indirect/social aggression when children develop

social relationships and social skills. Therefore, more socially developed children may generate

very different response options to an aggressive situation than less socially developed children.

In sum, children seem to select the response options that best suit their own abilities:

socially maladjusted children select aggressive responses that do not rely on social skills or the

social network because they do not possess these tools whereas socially skilled children select

aggressive responses that maximize their own capital: a social network and skilled social

interaction. Finally, Crick and Dodge's position assumes that socially adjusted children perceive










the nonaggressive response alternative (if they generate such an alternative as a behavioral

option) as more attractive and better suited to attaining their goals. For at least some people, a

nonaggressive response option may be seen as less effective, and selection of such an option

would be a mistake.

Self-Presentation and Reputation in Aggression's Aftermath

In traditional aggression experiments, participants are given the opportunity to aggress

against other participants or confederates-virtual strangers. While true that people do aggress

against those they do not know (e.g., rape and murder victims picked at random, mob aggression,

etc.), most violence does not occur between total strangers. For example, in 2005, the U.S.

Bureau of Justice estimates that seven out of ten rape victims knew their attacker and in 2002, of

solved murders, 75% of the victims knew their attacker (Bureau of Justice Statistics Crime

Characteristics). Although there are certainly some aggression processes that would be

unaffected by aggressing against known others, aggressing against others who one must face

later introduces a new problem. When one is motivated to harm another but must face the person

in the future, it is worth avoiding being perceived as an aggressive individual. Being perceived as

nonaggressive maintains positive relationships and the element of surprise for later aggression.

This conundrum of desiring a nonaggressive image yet desiring aggressive behavior can be

thought of in terms of the effect/danger ratio. Bj rokqvist and his colleagues (1994) state that,

"the effect/danger ratio is an expression of the subj ective estimation of the likely consequences

of an aggressive act. The aggressor assesses the relation between a) the effect of the intended

strategy, and b) the dangers involved, physical, psychological or social, for him/herself, and for

people important to him/her. The obj ective is to find a technique that will be effective and, at the

same time, incur as little danger as possible" (p. 28-29). In other words, when individuals

aggress against known others or those that they may encounter again in the future, there is the









added concern of prolonged repercussions for their actions and the danger component of the

effect/danger ratio is heightened.

In traditional studies, the fear of future repercussions is minimized because the victim is

someone that participants have never met and with whom participants have no reason to expect a

future encounter. Neglecting to consider this aspect of interpersonal aggression ignores the

fundamental human concern to maintain a particular image--images that rarely include being

aggressive. Although there is a desire not to be seen as aggressive (particularly by those whom

people must regularly encounter), the occasional desire to aggress against known others still

exists. Therefore, people face a conundrum: how does one aggress against others while

maintaining a nonaggressive image in the eyes of others? One possibility is using aggressive

means that heighten the effectiveness of the danger to compensate for the added danger of future

interactions with one's victim. Another possibility is to reduce the danger component through the

use of skillful self-presentation tactics.

Self-presentation is the tool that people use to ensure that their desired images are

maintained (Schlenker, 1980; Schlenker, 2003). These specific images may vary from situation

to situation and may be communicated through a variety of techniques such as mannerisms,

anecdotes told, manner of dress, manner of speak, props, etc. Although self-presentation does not

necessitate deceitfulness, it certainly can be deceitful (such as claiming an identity that in no way

reflects a person' s actual image). Self-presentation also incorporates exaggerations of naturally

occurring tendencies such as being friendly, reserved, generous, caring, articulate, etc., and

omissions or underplaying other qualities. Any particular image desired to be presented is

motivated by the specific social situation although some images may be universally desired. All

self-presentations are restricted by reality and the believability of any given image claim: a










person asserting a particular identity can successfully do so only to the extent that the witnessing

audience possesses no information conflicting with this identity. Keeping one's audience in the

dark can be established by the audience a) possessing no information bearing on this claim

(neither supporting nor conflicting with the desired image) or b) possessing information that is at

least somewhat consistent with the desired image. In an aggressive situation, a person faces a

difficult problem because his/her aggressive actions and traditionally desired nonaggressive

identity are inconsistent. In order to successfully portray the desired image, one must make this

inconsistency disappear. Specifically, the person must consciously disguise his/her actions as

nonaggressive by creating enough ambiguity that the victim or other audience a) must draw a

nonaggressive conclusion or b) at least consider a nonaggression interpretation a viable option.

Hardin and Higgins (1996) argue that all actors in the social environment simultaneously

negotiate a shared reality; together, actors provide explanations and interpretations of their

world, and reality is defined as the interpretation that all actors mutually agree upon. Actors are

then obligated to treat this negotiated truth as truth. However, as actors encounter new situations

and motivations, they may desire to renegotiate reality or find it necessary to defend the current

reality. For example, perhaps a particular person previously considered a suitable friend commits

a transgression serious enough to warrant being recast as an unsuitable friendship partner, or a

reality in which a particular woman is considered nonaggressive is threatened by her aggressive

actions. As such, individuals may provide alternative realities (e.g., recasting the role of the now

unsuitable friend) or bolster the believability of the current negotiated reality by providing

alternative explanations of the woman' s behavior. The ability to manipulate the situation and the

conclusions that can be drawn is a valuable skill.









Teasing provides one example of covert aggression. In many interpersonal relationships,

teasing is considered a sign of affection and a way to demonstrate and increase social

cohesiveness, although an abundance of research treats teasing as an aversive experience

contributing to long-term maladjustment (e.g., Kowalski, Howerton, & McKenzie, 2001). As

Campos, Keltner, Beck, Gonzaga, and John (2007) point out, "Teases, then, are ambiguous

communications that convey both negative and affiliative intent....Given this ambiguity, the

relational consequences of a tease depend on the extent to which the target interprets the tease as

an affiliative gesture that is an affront to the self' (p. 4). In personal relationships characterized

by teasing, the interpersonal partners may attribute any seemingly adverse teasing as benign.

Indeed, research supports the idea that satisfied romantic partners will adjust attributions of their

partners to maintain a positive view of the partner and relationship-excusing negative behavior

for those they like (Bradbury & Fincham, 1992; Goldstein, Tisak, Persson, & Boxer, 2006).

However, this tendency may be abused when a relationship partner becomes upset with his/her

significant other (or with one's friend) and lashes out verbally at one's partner. If confronted for

this hurtful behavior, an aggressor can dismiss the aggressive explanation by claiming that he/she

was "only teasing" and rely on the actual ambiguity of teasing and the general positive

attributions in the relationship to cover one's true intentions.

Teasing is only one of several methods that individuals may use to aggress that easily

lend themselves to alternative explanations in certain situations. For example, aggressors who

choose to ostracize a particular person can easily deny that the ostracism is intentional but rather

merely accidental (the aggressors have plans with others, the victim is paranoid, etc.). Similarly,

gossiping about a particular person can be framed as venting without malicious intent or as

seeking advice. The use of double entendres may also provide a safe route to express negativity









that provides two plausible meanings--one innocuous meaning and another more risque,

offensive, or combative meaning. Other methods may include failing to aid a friend when aid is

expected. For example, just as people try to manage the impressions that they themselves make

on others, people also tend to help their friends make the impressions that will serve their

friends' best interests, thereby engaging in beneficial impression management (Schlenker &

Britt, 1999). Consider a person who desires to impress a potential date or employer. He/she may

ask a mutual friend to help create an image in the eyes of the potential date as an attractive and

fun individual or as a competent and intelligent individual to the potential employer. The mutual

friend's failure to do so may be perceived as an act of aggression.

The primary emphasis in social psychology is how individuals behave and respond to a

social situation, frequently resulting in the characterization of people as responding to the

situation around them. This approach possesses a great deal of explanatory power but even more

explanatory power can be gained by increased attention to the volition and decisiveness of

individuals. Some research has demonstrated that people can and do actively construct their

situation to influence the attributions that audiences draw about them and their behavior. One

particularly well-known example is self-handicapping through which individuals negate

unflattering dispositional attributions in the event of failure outcomes while simultaneously

augmenting flattering dispositional attributions in the event of successful outcomes (Berglas &

Jones, 1978). The construction of an actual performance impediment provides an alternative

explanation for any forthcoming failure.

This type of behavior indicates that people are clearly sensitive to the attributional

tendencies of others and will actively try to control these attributions. Although constructing

their own alternative attributions (via excuses, self-handicaps, lies, etc.) may be preferred,









individuals can also sophisticatedly capitalize on existing situations to ensure the most favorable

attribution. One classic example of this ability is illustrated by Snyder, Kleck, Strenta, and

Mentzer (1979) in which participants tried to control attributions that might suggest that they

possessed negative attitudes toward physically handicapped individuals. Specifically,

participants were told to select a seat to watch a movie, choosing between watching a movie

while sitting beside a person with a visible handicap (wearing a leg brace) or watching a movie

in a different room while sitting beside a person without the visible handicap.

In one condition, the person's handicap was the only difference between the two rooms--

the same movie was playing in the two separate rooms. However, in another condition, the

rooms differed both in the person already seated and the movie playing. Therefore, in the former

condition, participants choosing one room over the other might reasonably be perceived as

choosing one room over the other based solely on the presence or absence of a handicap of the

person already watching the movie. As such, choosing the room without the handicapped person

would likely be considered an attempt to avoid the handicapped person. In the latter condition,

however, choosing to watch the movie without the handicapped person could be attributed to

either avoiding the handicapped person or a desire to watch the different movie in the other room

(because both movie and confederate differed between the two rooms). Participants almost

exclusively sat with the handicapped confederate when there was no plausible alternative

explanation (same movie showing in both rooms). However, when both the movies and the

confederates differed between the two rooms (providing attributional ambiguity), participants did

not seek out the handicapped individual--choosing to sit with the non-handicapped person could

easily be attributed to the movie selection rather than disliking the handicapped person. In other









words, participants were aware of the conclusions that audiences would draw about their

intentions based upon their actions and factors present in the environment.

In a similar manner, Gaertner and Dovidio (1977) demonstrated that white participants

were less likely to help black participants if failing to help could be attributed to something other

than their own prejudice. In this study, white participants did not differ in their willingness to

help white versus black confederates if participants believed that they were the only ones who

could offer help. In contrast, white participants were significantly less likely to help black

confederates when they believed that other people could offer help. In other words, when white

participants have only themselves to blame for leaving a confederate unaided, they offered help

regardless of race. However, if participants' failure to help could just as easily be attributed to

another reason (i.e., others could help the person in need), then participants were significantly

less likely to aid the black confederate than the white confederate. In short, the participants acted

in accord with their socially proscribed prejudice when they believed it would not be easily

detected, but restrained from acting on this prejudice if their actions clearly identified their

prejudice beliefs.

Together, these two studies (Gaertner & Dovidio, 1977; Snyder, Kleck, Strenta, &

Mentzer, 1979) along with a host of additional research suggest that individuals are clearly aware

of the attributions that others are likely to make and are motivated to control these attributions to

preserve a desirable image in the eyes of others. This awareness of the attributional tendencies of

others may curb the use of aggression in situations that enhance conclusions that the individual is

acting in a deliberately hostile manner. Instrumental aggression in particular, even when enacted

covertly, may lead observers to deduce that the individual is acting in a deliberately hostile

manner rather than in a nondeliberate manner. Hostile/emotional aggression is carried out purely









for its own sake--aggression that is instigated by emotion such as anger, frustration, etc., and

carries with it some personal gratification. In contrast, instrumental aggression is carried out as a

means to some other end (be it good or bad in nature). Of course, as Bushman and Anderson

(2001) point out, aggression may be multiply motivated: aggression may be instrumentally

motivated, but still provide a measure of emotional satisfaction, if nothing more than a reduction

in physiological arousal after aggressing (Geen, Stonner, & Shope, 1975).

As is frequently the case when classifying behavior, Bushman and Anderson (2001) argue

that emotional vs. instrumental aggression is best characterized as a continuum from exclusively

instrumental to exclusively hostile/emotional rather than a dichotomy. Likewise, covert

aggression may serve multiple purposes. In keeping with the notion that social actors knowingly

manipulate the attributions that others make for their behaviors, social actors may be especially

guarded in their aggressive interpersonal actions when their victims or other observers are

particularly likely to interpret behavior as selfish. In other words, when a person clearly gains

from an action (as is the case in instrumental aggression), others may be especially suspect of

this person. In short, instrumental covert aggression may be more desirous, but will be less likely

than emotional covert aggression, to the extent that the person's gained rewards are evident to

others. It is my contention that this type of behavior occurs regularly amongst those motivated to

aggress against others. Individuals weigh the risks of getting away with such behavior against the

cost of acting on their proscribed desires to inflict harm, respond aggressively, etc.

Situational Factors

Several factors can systematically affect an analysis of the cost-benefit ratio of social

behavior. First, some groups and cultures proscribe openly expressing grievances more than

others. For example, Cohen and colleagues demonstrated that U. S. southerners are more likely to

suppress external indicators of anger longer than U.S. northerners (Cohen, Vandello, Puente, &









Rantilla, 1999). Some groups value openly expressing grievances as a testimony to honesty and

genuineness, and view denials of animosity and covert aggression as fake, hypocritical and

dangerous to the social fabric because individuals do not possess an accurate understanding of

which individuals they can truly trust. In contrast, other groups and cultures have established

norms that stress playing nicely, maintaining a facade of pleasantry and civility and perceive

open expressions of animosity as threatening, rude, uncivil, and/or immature (Cohen et al.,

1999). Although groups that permit and encourage the expression of grievances and anger

toward others may show higher absolute levels of clear aggression and nastiness, there is no

necessary reason to assume that the aggressive and negative motives differ across these two types

of groups. Presuming that such motives do not differ as a function of normative grievance

expression, individuals with negative motives in the civility culture must resort to more covert

means of expressing their negativity.

Beyond cultural proscriptions against expressing grievances to others within one's

culture/group, there may be both cultural and practical concerns when expressing grievances

against specific others such as those higher in status or authority. Particularly authoritarian

cultures may pressure group members from expressing hostility toward others higher in authority

while simultaneously increasing the likelihood that those of lower status will face repercussions

of hostility. In addition to these cultural and normative concerns, however, there exists the very

real possibility that those higher in status generally have greater ability to extract revenge for

expressed hostility/aggression than those lower in status. Therefore, if individuals are motivated

to aggress against a higher status person, the more cautious choice would be to aggress covertly.

Indeed, aggressing openly can be quite foolhardy, such as openly threatening one's employer.









In addition to the status differential between a potential aggressor and his/her victim, there

are other important factors that may mitigate the path chosen when facing aggressive motives.

For example, some relationships may be marked by a generally tension-free nature with the rare

occurrence of a desire to aggress or act on hostility. However, other relationships may be fairly

tempestuous but their long-term survival is desired or necessary. For example, romantic

relationships may be particularly tumultuous but still be valued and so openly aggressive

behaviors may be undesirable. On the other hand, the relationship may not be valued in its own

right, but a minimal level of civility must be maintained as a means to some other end; e.g., an

employee may not care for his/her boss but must maintain a minimally positive relationship with

the boss in order to ensure continued livelihood. Both of these types of relationships should be

marked by more covert than overt aggression.

Hypotheses

First and foremost, future interaction is a critical antecedent of covert aggression. Without

future interaction, self-presentational concerns are reduced. When people do not expect a future

interaction with someone else, there is no need to hedge bets by covertly aggressing. Instead, any

desire to aggress will be expressed overtly or stifled. Conditions in which participants do not

anticipate a future interaction where their protagonist/victim parallel traditional aggression

research and so little would be gained from investigating this possibility.

In strategic covert aggression, a person must carefully balance two primary concerns: a

desire to effectively harm and also a desire to maintain a nonaggressive image. Maintaining a

falsely pleasant facade encourages recipients of these covertly aggressive actions to perceive the

entirety of the communication as pleasant rather than unpleasant, thereby serving a dual purpose.

First, if the recipient of a covertly aggressive act is suspicious that the action is aggression, the

ambiguous nature ensures a reasonable chance that the perpetrator will evade retribution for a










crime that was doubtfully committed in the first place. Second, a person may act in a covertly

aggressive manner in response to a potential affront. If one later learns that no affront was

actually intended and the act was misperceived, one has a ready alternative explanation for one's

covertly aggressive behavior. In short, covert aggression serves a person well when he/she is

motivated to do harm to another person but prevent detection, and covert aggression also serves

as a strategic response to ambiguous communication from others. Responding in an ambiguous

manner allows individuals the opportunity to play both sides of the fence until more information

is gathered about the social situation. Until more information is available, a dually aggressive

and pleasant communication allows both the maintenance of positive interaction and negative

interaction simultaneously. Once more information is gathered, individuals can bolster the

appropriate interpretation (pleasant or unpleasant) through less ambiguous communication. Thus,

covert aggression may initiate an unpleasant interaction or it may be used a cautious response to

the ambiguous behavior of others.

Because of the subtle duality of a covertly aggressive communication and based on

previous research regarding indirect aggression (a similar but distinct concept), one would expect

that socially savvy/sensitive individuals would be particularly likely to employ this strategy both

of their own volition and also in response to communications from others. However, when

people expect no future interaction with someone, there is no need for covert communication or

hedging bets because there exist few (if any) repercussions for overt aggression. In situations that

elevate consequences for acting overtly aggressively, individuals should be more likely to choose

covert aggression over overt aggression. This leads to Hypothesis I:

*Hypothesis I: The likelihood of covert aggression is heightened when future interaction is
expected, particularly when one is partially or wholly dependent on the other in that future
interaction.









Moreover, the presence of a covert aggressive path should heighten the likelihood of

aggressing than when individuals' only recourse is open and direct aggression. This leads to

Hypothesis II:

*Hypothesis II: Overall aggression (combined straight forward and ambiguous aggression)
will be less when one is forced to aggress openly than when the opportunity to aggress
covertly is also available.

If no covertly aggressive option is available, individuals particularly concerned about the

consequences for aggressing at all will be less likely to aggress in any manner--overtly or

covertly. The consequences for aggressing at all are heightened to the extent that future

interaction is likely or imminent, and to the extent that there are strong proscriptions against

hostile behavior. Such proscriptions can arise due to cultural climate or status differentials (i.e.,

aggressing against individuals higher in power).

Specific types of individuals may be especially likely to engage in covert aggression.

Because covert aggression and indirect aggression may involve the same types of behaviors,

scales assessing the tendency to engage in indirect aggression may also assess predilection to

engage in covert aggression. Several different measures of aggressive tendencies have been

developed to assess predispositions toward aggression and aggressive personal histories. Of

course, several scales focus on more overt forms of aggression, but some scales assess both

proclivity to engage in direct overt forms of aggression as well as subscales tapping emotions

that lead to the aggressive motive. For example, the Buss-Perry Aggression Questionnaire (Buss

& Perry, 1992) does not exactly assess indirect aggression, but it does include subscales

assessing not only physical and verbal aggression but also hostility and anger. In particular, later

researchers demonstrated that the hostility subscale predicted indirect aggression better than the

other subscale components and thus may be a useful tool in predicting whom is likely motivated

to covertly aggress against others (Archer & Webb, 2006).









Other scales have come closer to assessing covert aggression. Richardson's Conflict

Response Questionnaire (Richardson & Green, 2003) is a 28-item scale that assesses the

tendency for individuals to use various strategies under conflict. The two sub scales consist of a

score for both direct (e.g., frequency of pushing, grabbing, or shoving someone; frequency of

yelling or screaming at someone) and indirect aggression (frequency of spreading rumors about

someone; frequency of telling others not to associate with someone). Other popular methods of

assessing indirect aggression utilize peer nomination and/or ratings of all members of a specific

group (frequently all students in a class) (Bjoirkqvist et al., 1992). However, although this

technique may work well with young children who have a limited and structured social world

easily amenable to psychological research, this method is harder to employ with adults whose

social worlds are not as easily compartmentalized into a single classroom. In the current study,

both Richardson's Conflict Response Questionnaire and the Buss-Perry Aggression

Questionnaire will be assessed as potentially moderating variables to predict likelihood of

engaging in covert aggression and to test Hypothesis III and IV:

* Hypothesis III: Participants with greater proclivity toward (1) hostility, (2) anger, (3)
verbal aggression, and (4) indirect aggression will show higher rates of covert aggression.

* Hypothesis IV: Participants with greater proclivity toward direct aggression will show
higher rates of open aggression.

It is possible that individuals low in Hostility (as assessed by the Buss-Perry Aggression

Questionnaire) or low in Indirect Aggression (as assessed by the Richardson's Conflict Response

Questionnaire) may be less likely to (a) interpret an ambiguously aggressive response as

aggressive and (b) employ ambiguous aggression in response.

In addition to measures assessing aspects of aggression, increased empathy (particularly

the perspective taking aspect of empathy) actually provides individuals with increased

knowledge of their victim and ability to cause harm to their victims. Therefore, the perspective









taking and empathic concern subscales of Davis's Empathy Scale (1980) should also be assessed

to assess Hypothesis V:

* Hypothesis V: Individuals high in (1) empathy, especially(2) perspective taking, will show
increased likelihood of using covert aggression.

Finally, despite the covert measures by which participants will be able to aggress in the

current study, some participants may be particularly reluctant to act in a harmful manner, despite

the temptation. Specifically, participants high in need for approval should be less likely to

engage in aggression of any sort, including covert aggression. To account for this possibility, the

Crowne-Marlowe measure of Social Desirability (Crowne & Marlowe, 1960) will be included in

the current study. Thus, Hypothesis VI:

* Hypothesis VI: (1) Individuals high in social desirability will be less likely to engage in
aggression of any sort (either covert or open), but when they do, (2) they should show a
greater preference for covert aggression than their low social desirability counterparts.

In short, the following predictions are made about each condition's aggressive behavior:

* Hypothesis VII: (1) Individuals receiving positive feedback will respond in kind to their
evaluator--selecting positive messages. (2) The level of positive and negative messages
sent will not differ between positive feedback participants who have a covert method of
aggression and those who do not have a covert method of aggression available.

* Hypothesis VIII: (1) For participants expecting an interaction, individuals receiving
straightforward negative feedback who do not have a covert method of aggression
available to them will send primarily neutral messages and few negative messages. (2) For
the negative feedback participants expecting an interaction but who do have a covert
method of aggression will choose to send primarily double entendre messages (covert
aggression), and few negative messages. (3) Negative- feedback participants with a covert
strategy available to them but who are not expecting an interaction will send primarily
negative messages and few (if any) double entendres (covert aggression).

* Hypothesis IX: (1) In general, participants who receive double entendre feedback (covert
aggression), will use more double entendres than participants who receive positive or
negative feedback. (2) More specifically, the participants who expect an interaction will
use more covert aggression than the participants not expecting an interaction, who will use
primarily negative statements with few double entendres.(3) Of those expecting an
interaction, participants with covert aggression available to them will use many double
entendres and few if any negative messages whereas (4) participants without covert
aggression available to them will send primarily neutral and positive statements.









Moreover, in addition to actual behavior and messages sent to others (i.e., the actual act of

aggression), beliefs about the other person will also likely be affected. These beliefs are unlikely

to be affected by the response options available to the participants. As such, the following

predictions are made regarding the beliefs about the original communicator.

* Hypothesis X: Communicators sending positive feedback will be viewed most favorably
by the participants in terms of (1) being liked the most and (2) participants expecting to be
liked by the communicator. Little difference is expected on these dimensions between the
communicator sending negative feedback and the communicator sending covert aggression
(double entendre) feedback.

Above and beyond the liking dimension and warmth of feelings toward the original

communicator, the participants should differ in their estimation of the communicators'

communication skills and deserved respect. Although being a recipient of covert aggression is

anticipated to be an unpleasant experience and should manifest into dislike for the

communicator, there may be admiration and respect for the skills necessary to successfully

implement such a covertly aggressive maneuver. Therefore, the following is predicted:

* Hypothesis XI: Communicators sending double entendre feedback will be seen as more
skillful communicators than both (1) negative feedback communicators and (2) positive
feedback communicators. (3) Positive feedback communicators should be perceived as no
more skillful than the negative feedback communicators, but a halo effect may lead to a
slightly higher evaluation of the positive feedback communicator' s skillfulness than the
negative feedback communicator's skillfulness.

As suggested by research on the self-serving bias, individuals who receive negative and

ambiguous feedback should diminish the validity of the test to downplay the relevance of the test

in order to minimize the self-implications of negative feedback. As such, recipients of negative

and double entendre feedback should not internalize the feedback they receive regarding their

own creative intelligence.

* Hypothesis XII: Participants receiving (1) negative and (2) double entendre feedback will
view the test as a less valid test of creative intelligence than the participants who receive
positive feedback.










* Hypothesis XIII: Participants receiving negative and double entendre feedback should not
internalize the negative feedback, although negative feedback participants will be more
likely to internalize the feedback than ambiguous feedback participants due to the possible
positive interpretation in the ambiguous feedback.

The emotional state of participants should also vary as a result of the feedback that they

receive. Specifically, emotions should map onto the interpretations of the feedback. These

emotions and anxiety should motivate participants to escape face-to-face interaction when

possible. Therefore:

* Hypothesis XIV: (1) Participants receiving positive feedback should report positive
emotions, (2) participants receiving negative feedback should report negative emotions,
and (3) participants receiving double entendre feedback should report both positive and
negative emotions.

* Hypothesis XV: Participants receiving negative and double entendre feedback will be
most likely to avoid face-to-face interactions and instead seek out less immediate forms of
communication (i.e., computer mediated interaction and intercom mediated interaction).

In sum, then, it was expected, generally, that participants would respond in kind to the

feedback that they received. In other words, positive feedback participants would respond

primarily in a positive manner, negative feedback participants would respond in a negative

manner, and ambiguous feedback participants would respond via mixed messages when

available, and would eschew outright negativity for positive messages when mixed messages

were not available (and only neutral messages were). Likewise, the ability to employ mixed

messages should similarly curtail the use of negative feedback participants' endorsement of

outright negative messages. Moreover, a future interaction was expected to restrain the use

outright negative messages in favor of mixed messages. It was also expected that ambiguous

feedback and negative feedback participants would view their partners in a similar negative

manner relative to positive feedback participants, who were expected to view their partners

positively .









Current Study

The current study takes a unique approach to understanding interpersonal covert

aggression. Little psychological attention has been given to the facet of aggression that requires

continued interpersonal contact and thus a need to maintain an interpersonal identity as a

nonaggressive person. I contend that when individuals anticipate future interaction with another

person, the need for a costs/benefits analysis is heightened because in ongoing relationships, it is

even more important to maintain a nonaggressive interpersonal identity. As such, people will be

strongly motivated to avoid actions that will cause others to view them as aggressive rather than

nonaggressive. However, because the motivation to aggress can be particularly strong, people

will create opportunities or act on existing opportunities in the environment that capitalize on

attributional ambiguity to escape dispositional aggressive attributions. When another person is

conveying potentially aggressive or demeaning intent, the recipient of this message must

determine the true meaning behind the ambiguity and respond appropriately. If the message is

misinterpreted as aggressive or negatively toned and one responds in a defensive manner or in a

retaliatory manner, one unnecessarily escalates an interaction into an openly aggressive and

unpleasant situation that could have deleterious effects for the ongoing relationship between the

interactants. In contrast, if one incorrectly misinterprets the communication as innocent, one may

face the consequences of trusting those who should not be trusted, failing to prepare against

future less covertly aggressive interactions, be perceived as foolish and naive, etc. Therefore,

when given the opportunity to rely on attributional ambiguity (e.g., through the use of double

entendres), people should be more likely to demonstrate aggressiveness than when no such

attributional ambiguity exists.

The extant research on aggression lacks experimental investigation of the process by which

individuals decide to engage in covert aggression, their success in using it, and the repercussions









for themselves and others. In addition, although some research suggests that individuals who

utilize indirect aggression (and possibly more specifically covert aggression) are more successful

interpersonally than those who do not use this strategy, the reason is unclear. The reason indirect

aggressors (and possibly covert aggressors) tend to have higher social status could be that a) the

use of covert aggression actually increases social status by inspiring respect in others, or b)

socially successful individuals are more adept at covert aggression but employing covert

aggression does not actively influence one's social standing. The current study is an

experimental analysis of this question.









CHAPTER 2
METHOD

Overview

The current study assessed reactions to varying types of communication: positive,

negative, and ambiguous evaluations with potentially strong positive or negative connotations.

These evaluations came in the guise of evaluative feedback regarding participants' attempts to

complete a creative intelligence test. Most participants (with the exception of those in two offset

conditions) expected a future interaction with the other participant who provided the evaluative

feedback. Participants then had the opportunity to send a return communication to the other

participant in response to the evaluative feedback. Some participants were then provided with a

way to aggress covertly whereas others did not have this path available to them, and instead were

only able to respond in a directly positive, negative, or neutral manner. Participants also reported

their perceptions of the other person to the experimenter and then selected what method of

communication (face-to-face, speaker system or computer mediated communication) they

wished to use for the secondary group-based creative intelligence assessment. The

communication method served as measure of how much participants chose to distance

themselves from the other participant.

Design

The design was a 3 (Feedback: explicitly positive, explicitly negative or ambiguous

overture) X 2 strategy (Strategy: mixed messages vs. neutral message response option) factorial.

Two offset groups that did not expect any interaction underwent the critical manipulation

receiving either negative or ambiguous feedback with an opportunity to respond via mixed

messages. These offset conditions were included, pending variations in the use of mixed

messages, to test the importance of anticipating an upcoming interaction.









Participants

Participants (N = 295) who reported English as their primary language were recruited from

introductory psychology courses. Of these, 68 participants (54 females and 14 males) were

randomly assigned to the offset comparison conditions leaving 227 participants in the main study

design (170 females and 57 males). Participants were given credit toward completion of their

course research requirement.

Prescreening Session

Prior to arriving at the main experimental session, participants participated in a brief

prescreening session to gather individual difference information. As such, they completed the

Richardson Conflict Response Questionnaire (Richardson & Green, 2003), the Buss-Perry

aggression Questionnaire (1992), and the Social Desirability Scale (Crowne & Marlowe, 1960).

To enhance the cover story, participants also completed some filler questions tapping

extraversion, conscientiousness, and openness to experience. See Appendix A.

Procedure

Participants were run individually. When participants arrived, they were greeted by an

experimenter, seated in a room by themselves, and asked to read and sign an informed consent,

read a Study Overview briefly describing the purpose of the study. Participants were told that the

study was investigating group problem solving versus individual problem solving efficacy

throughout early adulthood. As such, participants first completed a creative problem solving

assessment by themselves that ostensibly served as a comparison baseline to a second creative

problem solving assessment completed collaboratively with another person. Prior to completing

the first assessment, participants completed the perspective taking and empathic concern

sub scales of the Davis Empathy Scale (Davis, 1980). Participants were also asked to briefly









describe why they were attending the University of Florida, their future plans, and what they

enj oy doing for fun on the Basic Information Questionnaire. See Appendix B.

Participants were told that the individual version of their creative problem solving

assessment asked them to design a squirrel-proof birdfeeder and that their fellow worker was in a

separate room working on a different form of the same test (a moon crash landing survival

problem). To do so, participants were given six minutes to design, diagram, and explain a

birdfeeder design that would prevent squirrels from accessing birdseed but still allowed birds

access. Participants were given several sheets of blank paper to complete their design.

After completing the first assessment, the experimenter returned and collected the

participants' designs. The participants were reminded that in many real-life work situations,

people frequently receive feedback from co-workers that can be helpful when later tackling a

similar problem and that such information can be elucidating. The current study was attempting

to mimic that tendency in the real world by allowing the two individuals to evaluate and

comment on each others' initial assessment prior to working together on second assessment.

Participants were presented with the results of the other participant' s (the fellow worker) creative

problem solving assessment. Again, under the guise of mimicking real-world interaction in

which individuals must provide evaluations of their peers' work, participants were asked to

evaluate the quality of the other participant' s solution to his/her problem via a free response item.

The other participant's solution to the problem was a reasonably good but imperfect solution to

the other participant' s own problem. While the participant was evaluating the fellow worker' s

solution, the fellow worker was supposedly evaluating the participant' s own drawing, and after a

few minutes, the experimenter returned to retrieve the participant's evaluation before exchanging

them with the evaluation of the participants' own drawing by the fellow worker and the fellow









worker' s Basic Information Questionnaire (see Appendix C). The Basic Information

Questionnaire conveyed a basic level of information about the fellow worker and indicated the

other participant was the same gender as the participant himself/herself.

Participants anticipating an interaction were randomly assigned to either the positive,

negative, or ambiguous feedback condition. The positive feedback indicated the fellow worker

perceived the participants' designs as generally good, as showing a good deal of potential, and

praising the participants' creative intelligence. The negative feedback indicated that the fellow

worker perceived the participants' designs as faulty and uncreative, and disparaged the

participants' creative intelligence. The ambiguous feedback conveyed two possible

interpretations to the participants: both a potentially positive meaning and a potentially negative

meaning. The ambiguous feedback was pilot tested to determine that, given the nature of the

task, participants perceived the feedback to convey these two potential meanings. (See Appendix

D). Participants assigned to one of the two offset conditions not expecting an interaction were

assigned to receive either negative or ambiguous feedback, but not positive feedback as that was

not a primary concern for the current study.

After reviewing the Basic Information Questionnaire and reviewing the fellow worker' s

feedback, the participants were then asked to communicate their initial impressions of the person

to both the fellow worker himself/herself via the First Impressions Questionnaire and

subsequently to the experimenter via the Preliminary Interaction Assessment. The First

Impressions Questionnaire was supposedly sent to the fellow worker to facilitate the secondary

assessment in that it is helpful to know what other people believe about oneself. On this

questionnaire, participants were presented with a list of eighteen statements. Half of the

participants were randomly assigned to receive a checklist that contained three different types of









descriptors, positive, negative, and mixed message descriptors (conveying both a positive and

negative connotation). For example, a positive descriptor read, "This person is capable of doing

many things." A negative descriptor read, "This person finds novel tasks difficult." A mixed

message descriptor read, "The quality of this person' s work is well-known." (Appendix E.)

The other half of the participants were randomly assigned to receive a checklist that only

contains positive, negative, and neutral descriptors (no double entendres). (Appendix F.) In short,

the participants in this condition had no ambiguous path available to them by which to respond

safely--they had to choose to respond in a directly positive, negative, or neutral manner.

Selecting neutral descriptors allowed individuals to take a middle-of-the-road approach but did

not allow them the option of presenting a dually friendly and aggressive image. A neutral

approach presented neither of these images. The eighteen statements were equally divided across

the three types of responses presented to the participants (positive, negative and either neutral or

mixed messages). Participants were asked to check off up to 8 items that they believed accurately

described the person with whom they were interacting. After selecting 8 items, participants were

asked to rank these 8 statements from most descriptive to least descriptive of the other person.

This approach allowed participants to select all or none of the items in each category--if

participants wished to eschew double entendres entirely, they had that ability.

After participants sent this information, they were told that the next phase of the study was

the collaborative second assessment. Under the guise of assessing any extraneous variables that

could potentially affect performance on the collaborative assessment, participants completed the

Positive and Negative Affect Scale (Watson & Tellegen, 1988) described as part of a Preliminary

Interaction Assessment (see Appendix G), as emotional responses may have mediated the extent

to which participants derogate the fellow worker' s personality or ability.









The remainder of the Preliminary Interaction Assessment assessed the feelings that

participants had toward the other person, the impending interaction, and themselves such as how

much they anticipated liking the other person, anticipated being liked by the other person,

assessment of the other' s social skills, etc. This questionnaire gave the participants the

opportunity to express their negativity toward the fellow worker more openly because they were

told that this questionnaire would not be shown to the fellow worker. Moreover, it was stressed

that it was important to answer honestly to ensure validity of the final assessment. In addition to

the previously mentioned questions, this questionnaire also assessed how positively participants

believed they evaluated the other participants' initial assessment.

I predicted that because participants would not have received feedback on their own

performance when evaluating the performance of others, they would have no reason to assess the

performance of the other in a strongly negative way. Previous research shows that when

evaluating a specific other person, people tend to evaluate that person fairly positively (as above

average; Klar & Giladi, 1997). As such, participants' evaluations of the fellow worker' s initial

creative intelligence assessment would be fairly positive. The participants' evaluations of the

partner' s work were obj ectively assessed by independent raters at the conclusion of the study.

However, participants who received a negative or ambiguously aggressive evaluation may have

altered their retroactive evaluations of the evaluations they gave earlier by doing one of two

things. First, they may have inflated the positivity of their evaluations to bolster their justification

for a hostile reaction or, alternatively, they may have deflated the positivity of their evaluations

to portray themselves as less naive in their willingness to "play nice;" in short, participants may

have portrayed themselves as having more of an upper hand than they actually had. Questions









assessing this possibility were only exploratory in nature, as either of these two possibilities were

equally viable possibilities.

Finally, the experimenter told participants that there was a degree of flexibility regarding

what communication method the participants used to collaborate on the final assessment and

because the study had not been active for very long, the study still needed people to

communicate in all methods. Therefore, the experimenter offered the participant a choice. The

participant could choose between interacting face-to-face with the fellow worker, via a speaker

system (only audio transmission), or via a computer chatting program. This measure assessed the

degree to which participants desired control over their communications and psychological

distance from the communication with the fellow worker. After the participants made their

selection, the participants were carefully probed for suspicion, debriefed, and dismissed.









CHAPTER 3
RESULTS

Pilot Testing

Both the feedback and the statements in the first impressions questionnaire were pilot

tested to ensure that multiple meanings were able to be perceived. Several variations of

ambiguous feedback were tested, as were several variations of neutral and mixed messages. For

the ease of presentation, only the final selected measures will be discussed.

Feedback

Participants (N = 77) completed the squirrel proof birdfeeder problem described above.

Then, participants were asked to read one of the three possible versions of the feedback and

asked to (1) report their interpretation of the feedback, and (2) report, if such feedback had been

actually based on their birdfeeder solution, how they would feel about receiving such feedback.

Specifically, participants reported how positive an evaluation the feedback conveyed, how

negative an interpretation it conveyed, and how much a negative and positive interpretation were

simultaneously conveyed. Moreover, participants reported how sincere the author of such

feedback was, how upsetting the feedback would be, and how much the author of such feedback

was trying (1) to conceal an insult and (2) to be superior. All of these ratings were made on a

scale of 1-9 Likert-type scale. See Appendix D for the specific wording of the feedback and

dependent measures.

A significant effect of feedback was found for ratings of positivity (F (2, 74) = 126.39, p <

.0001, r12 = .774), negativity (F (2, 74) = 111.58, p < .0001, r12 = .751), and simultaneous

positive and negative meanings (F (2, 74) = 8.22, p = .0006, r1,2 = .182). See Table 3-1 for mean

ratings as a function of feedback. Specifically, although participants perceived the ambiguous

feedback to be as positive as the positive feedback, they perceived the ambiguous feedback to be









more negative than the positive feedback. In addition, the ambiguous feedback was perceived to

convey a double-meaning significantly more than both the positive and negative feedbacks.

Regarding visceral reactions and perceptions of the feedbacks' authors, participants

expected being most upset after receiving the negative feedback significantlyy more than both the

positive and ambiguous feedbacks), and ambiguous feedback was seen as more upsetting than

positive feedback. In keeping with this pattern, participants reported that the negative feedback

and ambiguous feedback authors were most trying to appear superior significantlyy more so than

the positive feedback author). Ambiguous feedback was seen as trying to conceal an insult

significantly more so than positive feedback; negative feedback was seen as intermediate and not

significantly different from either positive or ambiguous feedback. Likewise, the same pattern on

the inverse question was evident such that ambiguous feedback was seen as significantly less

sincere than positive feedback, while negative feedback was intermediate and not significantly

different from either.

In sum, ambiguous feedback was perceived as conveying both a negative and positive

meaning, was more negative than positive feedback and was more positive than negative

feedback. Likewise, ambiguous feedback was more upsetting and less sincere than positive

feedback, and was seen as more of an attempt to be superior and conceal an insult than positive

feedback. See Table 3-1 for mean ratings of positive, negative, and double-meaning

interpretation of each feedback as well as pilot participants' reactions to the feedback and the

feedback communicator. In short, the various forms feedback indicated that they had the desired

effect; the feedback was effective in inducing the desired perceptions in the pilot participants.

Message Options

For the First Impressions Questionnaire in which participants select and rank statements to

send to their partner after receiving feedback, it was necessary to have statements with a single










positive meaning, statements with a single negative meaning, statements with a simultaneously

strong positive and a strong negative meaning (mixed messages), and statements that were

relatively neutral-- i.e., a single low negative/slightly positive meaning. The same participants as

above also rated possible statements as to how positive they were, how negative they were, and

the extent to which they had both a positive and negative interpretation on a scale of 1 (not at all)

to 9 (very much). The following guidelines were used to select statements for each category.

Positive statements had to have a double-meaning score < 5, a positive score > 6.5, and a

negative score I 3. Negative statements had to have a double-meaning score < 5, a positive score

< 3, and a negative score > 6.5. Mixed message statements had to have a double-meaning score

>5, a positive score > 6, and a negative score > 4. Neutral statements required a double-meaning

score < 5, a positive score I 6 but > 3, and a negative score <4. See Appendices E and F for the

final selected statements that met these criteria. Thus, the messages from which the actual

participants could select were rated by the pilot participants as having the desired properties.

Scoring the First Impressions Questionnaire

To determine the use of positive, negative, or alternate responses (mixed messages or

neutral messages), the 18 statements of the First Impressions Questionnaire were divided into the

three categories: (1) 6 positive messages, (2) 6 negative messages, and (3) 6 alternate messages,

which consisted of either mixed messages (for participants in the mixed message condition) or 6

neutral messages(for participants in the neutral condition). Participants selected 8 total messages

and then ranked these 8 messages from most descriptive (rank = 1) to least descriptive (rank =

8).

Two types of information can be gathered from the First Impressions Questionnaire (FIQ).

First, what messages did the participants select? For each category, the number of messages

selected from each category was tallied such that if a participant selected 2 positive, 1 negative,










and 5 alternative messages, then he/she would have a positive-selection score of 2, a negative

selection-score of 1, and an alternative-selection score of 5.

The second piece of information to be gathered from the FIQ is how descriptive of their

partner the participants perceived the selected messages from each category to be (indicated by

the rank-value assigned to each statement). Therefore, rank-scores were computed by averaging

the ranks assigned to the 6 messages from each category. Because ten messages were not

selected and therefore not ranked, a rank value of 13.5 was assigned to these unselected 10

messages by computing the average value that would have been assigned to the unselected 10

messages if all 18 messages had been ranked. For example, if a participant selected 2 positive

messages and ranked them 1st and 2nd most descriptive and did not select the other four positive

messages, then this participant's positive-rank score is an average of 1, 2, 13.5, 13.5, 13.5, and

13.5. If none of the six messages from a given category were selected, then the score for that

category would be 13.5 as all messages in that category would be assigned a score of 13.5. If all

messages from a given category were selected and ranked as most descriptive, then the score for

that category would be 3.5 (i.e., an average of 1, 2, 3, 4, 5, and 6). Thus, the possible range of

scores for each category range from 3.5 to 13.5 with lower scores reflected greater endorsement

of that type of message.

These two measures of category endorsement were highly correlated with each other. As

would be expected, as more statements were selected from each category, the average rank for

each category was higher (where higher ranks are reflected by lower numerical values).

Specifically, selection scores and ranking score were highly correlated for each category: r attemnate

statements (220) = -.91; r negative statements (220) = -.99; r positive statements (220) = -.98; all p' s < .0001.

Due to the extremely high correlation between these selection and ranking measures, only










analyses regarding the average ranks assigned to each category are presented for ease of

presentation. The patterns of significant and non-significant Eindings were essentially identical

for the ranking and selection measures, and any exceptions are indicated below.

Overview of Analyses

In the analyses below, participants who anticipated an interaction with their partner (N=

227) were included in a series of 3 (Feedback: Positive, Negative, or Ambiguous) X 2 (Response

Option: Mixed Messages or Neutral Messages) general linear models that tested the use of

positive messages, negative messages, and alternative messages (mixed messages or neutral

messages, depending on response option condition). Although the effects on each dependent

measure are interesting in and of themselves, testing specific hypotheses requires comparing

across categories. Because the three categories are not independent of one another, a degrees of

freedom restriction occurs if all three categories are compared to each simultaneously. To test

hypotheses that predicted differences across categories, difference scores were calculated to

compare each of the three categories to the other two. These differences scores were then

submitted to the same general linear model described above. For ease of presentation, the

statistical significance of mean comparisons described below are presented in Tables 2 and 4.

Differences between specific conditions after a significant main effect involving more than 2

levels of a significant interaction were determined by tests of simple effects or specific contrasts.

The results are indicated via subscripts in the tables. Gender was included as a possible

moderator in each of the analyses, but failed to produce any significant interactions or main

effects and so was dropped from the analyses.

The offset comparison conditions were included in the current study to assess the extent to

which any differences in the use of mixed messages were affected by the possibility of a future

interaction or generalized to reactions regardless of future interactions. Therefore, the dependent









measures of the offset comparison conditions were assessed using the same statistical framework

with follow-up comparisons describing whether the future interaction mattered. These analyses

are described in more detail below.

Messages Used

Positive, Negative and Alternate Messages

Variations in the average rank-value of each category of messages (positive, negative, and

alternate) was determined for each combination of feedback and response option using the 3

(Feedback) X 2 (Response Option) general linear model approach described above on each of the

three message categories. The average ranks assigned to each category are presented in Table 3-

2. Means for the average number of statements selected from each category (the selection scores

described earlier) are presented in Appendices H and I.

In general, the hypotheses predicted that participants would respond to the feedback they

received in a similar manner such that positive feedback participants would send positive

messages, negative feedback participants would send mixed messages when they were available

and negative messages when mixed messages were not available, and ambiguous feedback

participants would send primarily mixed messages when they were available but would resort to

neutral and positive messages when mixed messages were not available. Further, it was predicted

that the absence of mixed messages as a communication alternative would have no effect on the

primarily positive messages sent by positive feedback participants.

Regarding the use of alternate messages (neutral or mixed messages), a significant main

effect of feedback (F (2, 214) = 12.46, p <.0001, r1,2 = 104) and simple effects comparisons

indicate that negative feedback participants endorsed the alternate messages (Mnegative feedback

8.54) more than either positive feedback participants and ambiguous feedback participants (p 's <

.0001,M2~positive feedback = 9.65,M2~ambiguous feedback = 9.60). Surprisingly, ambiguous feedback









participants and positive feedback participants did not differ from one another (p = .81). These

Endings are not consistent with the hypothesized contention that ambiguous feedback

participants would most strongly endorse mixed messages as negative feedback participants were

the most likely to endorse mixed messages/neutral messages and positive and ambiguous

feedback participants generally avoided them.

Negative messages were also submitted to the above model. Main effects of both feedback

(F (2, 214) = 123.32, p < .0001, r12 = 53 5) and option (F (1, 214) = 6. 18, p = .01, r12 = .028)

were qualified by a significant Feedback X Option interaction (F (2, 214) = 7.23, p = .0009, r1,2

=.063). In short, negative messages were most endorsed (via higher rankings) by negative

feedback participants, and practically ignored by all ambiguous and positive feedback

participants (regardless of response option condition), and negative feedback participants ranked

negative messages higher than both positive and ambiguous feedback participants in both

response option conditions (p's < .0001). Negative feedback participants were most likely to

endorse negative messages when they were in the neutral message condition (Mnegative feedback-

Mixed Messages = 10. 78, M~ negative feedback Neutral Messages = 9. 16, p <. 0001)-possilbly endorsing negative

messages here because they could not utilize the negative interpretation of mixed messages to

send aggressive messages. Ambiguous and positive feedback participants did not differentially

endorse negative messages across response option conditions (p' s > .84). This pattern of Eindings

is consistent with the hypotheses that negative feedback participants would most strongly

endorse negative messages relative to other feedback conditions, and also that negative feedback

participants would curtail their use of negative messages when mixed messages were available.

Regarding the ranking of positive statements, a feedback main effect (F (2, 214) = 126.93,

p < .0001), r12 = .543), qualified by a Feedback X Option interaction (F (2, 214) = 6.21, p =









.0024, r1,2 = .055) indicates that positive feedback participants and ambiguous feedback

participants endorsed positive messages considerably more than negative feedback participants,

regardless of response option (p' s <.0001), but these positive feedback and ambiguous feedback

participants endorsed positive messages equally (p Mixed Messages = .27, and p Neutral Messages = .24) .

However, while positive feedback participants were unaffected by response option (p = .94),

negative feedback participants in the mixed messages condition endorsed positive messages

more strongly than participants in the neutral messages condition (p = .01). In contrast, the

availability of mixed messages made ambiguous feedback participants weaken their endorsement

of positive messages compared to ambiguous feedback participants in the neutral messages

condition (p = .04). In sum, as predicted, positive feedback participants did employ primarily

positive messages and were unaffected by the availability of mixed messages, but ambiguous

feedback participants employed positive messages just as much as positive feedback participants,

which was not anticipated. However, ambiguous feedback participants did behave consistently

with the hypothesized notion that they would endorse positive messages more strongly in the

absence of mixed messages as a viable communication alternative, i.e., in the neutral messages

condition. In other words, when ambiguous feedback participants had neutral statements as their

alternative response option, they endorsed outright positive messages more strongly than when

mixed messages were available. In contrast, negative feedback participants gave weaker

endorsements of positive messages when mixed messages were available than when neutral

messages were available. In other words, negative feedback participants were nicer when mixed

messages were a communication outlet than when they were not.

Category Comparison

To compare the use of one category to another, three difference scores were calculated for

the difference in the average ranks assigned to each category (i.e., difference scores to compare









alternative messages to negative messages, alternative messages to positive messages, negative

messages to positive messages). This way, relative endorsement of categories can be compared

to see if participants endorse, for example, negative messages more or less than alternative

messages. These difference scores were then submitted to the 3 (Feedback) X 2 (Response

Option) general linear model described above. Table 3-2 presents average ranks and when

differences between categories were significantly different from one another. For expository

purposes, Table 3-3 presents difference scores discussed below.

It was predicted that positive messages would be endorsed more than negative or

alternative messages for positive feedback participants, regardless of response option. In

contrast, when mixed messages were available, it was expected that both negative feedback and

ambiguous feedback participants would endorse mixed messages most, with secondary

endorsements of negative statements for negative feedback participants and positive messages

for ambiguous feedback participants. In contrast, when neutral messages were the alternate form

of communication instead of mixed messages, it was hypothesized that negative feedback

participants would primarily endorse negative messages and ambiguous feedback participants

would primarily endorse positive and neutral statements.

In general, participants in the positive feedback and ambiguous feedback conditions

strongly endorsed positive messages, avoided negative messages, and ranked alternative

messages intermediately. This was true regardless of response option. However, negative

feedback participants' endorsement varied as a function of response option; mixed message

participants rated positive messages and mixed messages equally highly and as more descriptive

than negative messages. However, negative feedback participants in the neutral messages









condition were more openly aggressive by rating negative messages more highly than positive

messages (and as highly as neutral messages).

More specifically, significant main effects of feedback were found for all three difference

score comparisons (F negative to alternative (2, 214) = 25.66, p < .0001, rl2 = .193; F positive to attemnative (2,

214) = 75.93, p < .0001, rq2 = .415; F negative to positive (2, 214) = 150.71, p < .0001, rl2 = .585). A

significant main effect of option was also found for comparison of negative messages to

alternative messages (F (1, 214) = 8.08, p = .0049, r1,2 = .036). These effects were qualified by

significant Feedback X Option interactions for each of the three category comparison difference

scores, (Fnegative to alternative (2, 214) = 3.24, p = .04, rlp2 =.029; F positive to attemnative (2, 214) = 3.93, p

=.02, rlp2 = .035; F negative to positive (2, 214) = 7.68, p = .0006, rlp2 = .070)2. In Other words,

differences between endorsing negative messages vs. alternative messages, positive vs.

alternative messages, and positive vs. negative messages differed depending on feedback and

response option for each kind of category comparison. The significant orthogonal simple effects

comparisons are presented in Table 3-2.

Within each type of feedback and response option, participants consistently ranked

negative messages as significantly less descriptive than alternative messages, with the sole

exception that neutral message condition negative feedback participants did not differ in their use

of negative messages and neutral messages (p = .19). Participants in the positive feedback and

ambiguous feedback conditions (in both response conditions) ranked positive messages as

significantly more descriptive than either negative messages or alternative messages (p's <

.0001). However, negative feedback participants in the mixed message condition did not differ in


2Although the three-way Category X Feedback X Option interaction was significant for
ranking scores, the comparison of alternative statements to negative statements selection was
only marginally significant, F (2, 216) 2.63, p .0741, r12 .024)









their rankings of positive messages and mixed messages (p = .12); negative messages were

ranked as significantly less descriptive than both of these, p < .0001 and p = .01, respectively,

indicating that while these negative feedback participants endorsed negative statements more

than positive or ambiguous feedback participants, these participants still exhibited a slight

preference for mixed messages and positive messages. Meanwhile, negative feedback

participants in the neutral message condition ranked positive messages as less descriptive than

both neutral messages and negative messages (p < .0001 and p = .0045, respectively) indicating

that these participants alone were the most nasty in their messages.

This pattern of findings is partially consistent with the predicted effects. Specifically, as

predicted, positive feedback participants did endorse positive messages more than negative or

alternate messages, regardless of response option condition. The participants were motivated to

be nice and overwhelmingly were. In contrast, ambiguous feedback participants were expected to

use mixed messages more than either positive or negative messages in the mixed messages

condition, but they employed positive messages more than mixed messages or negative messages

in this condition. On the other hand, ambiguous feedback participants' use of primarily positive

messages in the neutral messages condition was largely consistent with the hypothesis that

participants would favor positive messages when mixed messages were not available. It was

expected that negative feedback participants in the mixed messages condition would capitalize

on the availability of mixed messages and endorse these more than negative or positive

messages. While negative feedback participants did endorse mixed messages more than negative

messages, they were not endorsed significantly more than positive messages in the mixed

messages condition, indicating that perhaps the negative feedback participants were torn between

positive messages and mixed messages. In contrast, it was hypothesized that because negative









feedback participants in the neutral messages condition could not satisfy their aggressive urges

via mixed messages, they would endorse negative statements more strongly than the negative

feedback participants with mixed messages available.

Summary of Message Use

The simple effects comparisons are presented in Table 3-2 for all of the above analyses.

Orthogonal simple effects comparisons are presented in the table, with means having at least one

factor in common and sharing a subscript not significantly different from one another.

It was expected, generally, that participants would respond in kind to the feedback that

they received. In other words, positive feedback participants would respond primarily in a

positive manner, negative feedback participants would respond in a negative manner, and

ambiguous feedback participants would respond via mixed messages when available, and would

eschew outright negativity for positive messages when mixed messages were not available and

instead, only neutral messages were. Likewise, the ability to employ mixed messages should

similarly curtail the use of negative feedback participants' endorsement of outright negative

messages.

In general, participants receiving negative feedback were considerably more aggressive

overtly and covertly, relative to participants receiving ambiguous and positive feedback, as

determined by the ranks assigned to negative messages and mixed messages. Ambiguous

feedback participants did not respond as aggressively as was expected-participants responded

quite similarly and in a generally positive fashion after receiving positive and ambiguous

feedback. However, while positive feedback participants exhibited low levels of overt and covert

aggression and were entirely unaffected by response option, ambiguous feedback participants

were somewhat affected by response option (mixed messages versus neutral messages as the

alternative message). As compared to their neutral message counterparts, ambiguous feedback










participants were less nice when they had mixed messages available to them, but negative

feedback participants were nicer when they had mixed messages available to them. Thus, while

Hypothesis IX. 1 predicted that ambiguous feedback participants would use more mixed

messages than positive feedback or negative feedback participants, this was not supported as

negative feedback participants were most likely to endorse mixed messages.

Because response option (mixed messages versus neutral message conditions) had no

effect on rankings of negative messages or alternate messages made by positive feedback

participants (p's > .90), Hypothesis VII.2 (that positive feedback participants would not differ in

their use of positive and negative messages across response options) was supported. Positive

feedback participants were always nice and unaggressive, regardless of whether they were in the

mixed message or neutral message condition.

Because negative feedback participants were generally nicer in the mixed messages

condition than the neutral messages condition, but the reverse was true for ambiguous feedback

participants, Hypothesis II (that overall aggression would be less when one is forced to aggress

openly than when the opportunity to aggress covertly is also available) received mixed support.

Negative feedback participants did not follow Hypothesis II: these participants were less

negative when they had mixed messages available to them and did not differ in their usage of

alternate statements. Ambiguous feedback participants, however, ranked mixed messages more

highly than neutral messages (attempting to be covertly aggressive) while they did not differ in

their use of negative messages.

Hypothesis IX.3 predicted that ambiguous feedback participants in the mixed messages

condition would use many mixed messages and few, if any, negative messages. In comparing the

use of negative messages to the alternate messages, participants in each of the feedback and










response option conditions ranked negative messages as less descriptive than the alternate

messages, except for negative feedback participants in the neutral messages condition, who did

not differ in their rankings of negative messages and neutral messages, thus supporting

Hypothesis IX.3. Likewise, Hypothesis IX.4 predicted that ambiguous feedback participants in

the neutral messages condition would send primarily neutral and positive messages, and received

qualified support. While positive and ambiguous feedback participants ranked mixed messages

and negative messages similarly, negative feedback participants ranked negative messages and

mixed messages more similarly than both the ambiguous feedback participants, and the positive

feedback participants. This effect stems primarily from negative feedback participants ranking

mixed messages higher than positive feedback or ambiguous feedback participants. In other

words, while ambiguous feedback and positive feedback participants demonstrated a clear

preference for mixed messages over negative messages, negative feedback participants showed

the same preference to a much lesser degree, seeing a considerably smaller distinction between

mixed messages and negative messages.

One interpretation of the negative feedback participants' behavior is that when negative

feedback participants did not have the opportunity to aggress covertly, they shifted to overt

aggression. In contrast, negative feedback participants who could aggress covertly did so and

reduced their level of overt aggression. This primary endorsement of positive messages by

positive feedback participants (rather than negative messages or mixed messages) supports

Hypothesis VII. 1 that participants receiving positive feedback would primarily endorse positive

messages. In contrast, Hypothesis VIII was not fully supported. Specifically, Hypothesis IX.4

predicted that negative feedback participants in the neutral messages condition would send

primarily neutral messages and few negative messages, and this was found not to be the case.









Instead, while the general trend was that neutral messages were endorsed more strongly than

negative messages by those in the neutral messages condition, this difference was not statistically

significant. However, Hypothesis VIII.2 predicted that negative feedback participants nI ithr mixed

messages available to them would primarily endorse mixed messages over than negative

messages, and this was indeed found to be the case.

In sum, ambiguous feedback and positive feedback participants responded similarly and in

a generally positive manner to their feedback such that they primarily endorsed positive

messages, followed by alternate messages, and most weakly endorsed negative messages, which

were essentially ignored. The only real difference between positive and ambiguous feedback

participants lies in the manner in which they were affected by the availability of mixed messages.

Specifically, positive feedback participants were not at all affected by their response option

condition, but ambiguous feedback participants endorsed mixed messages more than neutral

messages, possibly to send aggressive overtones, but also possibly to send positive intimations.

In other words, the availability of mixed messages seems to have made ambiguous feedback

participants slightly less nice, as further indicated by their weakened endorsement of positive

messages when mixed messages were available. In contrast, for negative feedback participants,

the availability of mixed messages resulted in less negative message endorsement and stronger

positive message endorsement than the availability of neutral messages.

Objective Assessment of Design and Evaluation Quality

It is reasonable to ask whether some of the effects might be due to variations in how well

participants completed their own design or how positively participants had originally evaluated

their partners. For example, participants who had evaluated their partner very positively and

received negative feedback or ambiguous feedback may have reacted more strongly to feedback

than participants who had not evaluated their partner so positively. Likewise, participants who










had developed a particularly exceptional squirrel-proof birdfeeder may have responded more

strongly to negative or ambiguous feedback than participants who did not create such an

exceptional design.

Two independent raters evaluated each participant' s squirrel proof birdfeeder design on

four different dimensions: 1) how much effort the participant exhibited in creating the design,

2) how creative the design was, 3) how effective the design would be at keeping squirrels out of

a birdfeeder, and 4) how practical the design was (intraclass correlation reliabilities, ICC (3,k): r

=.69, .66, .58, and .29, respectively). The two raters were assumed to be the entire population of

raters and the reliability is for the mean of their ratings (Shrout & Fleiss, 1979). The raters also

evaluated the written evaluation sent to the nonexistent partner of the partner' s own creative

intelligence assessment on two dimensions: 1) how positive the evaluation was, and 2) how

much effort the participant put into evaluating the partner's work (a = .78 and .78, respectively).

None of these measures significantly moderated the above findings.

Interaction Expectations

The manipulation of interaction expectations was successful. Participants who were told

they would be meeting their partner should have reported that they were going to meet someone

more than participants who were told they would not be meeting their partner. A between-

subj ects t-test revealed that this was the case, t (142) = 8.70, p < .0001, Mmeett= 1.99, MrNo meet

1.47.

Anticipating a future interaction was generally expected to restrain the use outright

negative messages in favor of mixed messages such that participants would shift from mixed

messages to negative messages when not facing a meeting such that outright negative messages

would be used less and mixed messages would be used more when a meeting was imminent than

when no meeting was anticipated. To assess the actual impact of expecting an important,









imminent collaborative interaction with their partner on the likelihood of using various forms of

communication (positive, covert aggression, straightforward aggression), the participants in the

offset comparison conditions were included in a 2 X 2 General Linear Model: Feedback

(Ambiguous vs. Negative) X Interaction Expectation (Meet vs. No Meet), looking at each

statement category ranking (positive, negative, mixed messages). All participants in this analysis

had mixed messages available to them.

A marginal Feedback X Expectation interaction was found for the ranking of the positive

messages (F (1, 137) = 2.72, p = .10). As shown in Table 3-4, negative feedback participants

more strongly endorsed positive messages if they anticipated a future interaction (F (1, 137) =

4.06, p = .0460, Mmet = 9.3 1, M~Nomeet = 10.3 0), whereas ambiguous feedback participants did

not differ in their ranking of positive messages as a function of anticipated interaction (F (1, 137)

= 0.04, p = .85). See Table 3-4 for mean rankings as a function of feedback and meeting

expectation. Rankings of mixed messages and negative messages were unaffected by interaction

expectation (p ambiguous feedback = .30, p negative feedback = .44), or an interaction between interaction

expectation and feedback (p ambiguous feedback = .16, p negative feedback = .47). The general lack of

impact of an expected meeting fails to support Hypothesis I. In addition, because negative

feedback participants not expecting a future interaction primarily endorsed mixed messages

rather than negative messages, Hypothesis VIII.3 was not supported, which predicted that few

mixed messages would be endorsed, and participants would instead rely on negative messages.

Finally, Hypothesis IX.2 predicted that ambiguous feedback participants not expecting an

interaction would use more mixed messages than those expecting an interaction and would

primarily endorse negative messages. However, because ambiguous feedback participants

expecting an interaction did not differ in their use of mixed messages compared to ambiguous










feedback participants not expecting an interaction and sent primarily positive messages rather

than negative messages, Hypothesis IX.2 was not supported.

Perceptions of Partner

Twenty-three questions from the Preliminary Interaction Assessment assessed

participants' perceptions of their partner and reactions to the test, and these questions were

grouped into nine factors including how likeable, smart, condescending, straightforward, sincere,

and a skilled communicator the other person was as well as how liked they expected to be by

their partner, how smart they considered themselves to be, and how valid they believed the test to

be. See Table 3-5 for a list of the items in each trait/grouping and relevant alphas. It was

generally predicted that ambiguous feedback participants not expecting an interaction would use

more mixed messages than those expecting an interaction and would primarily endorse negative

messages.

Main effects of feedback were found for how likeable straightforward, sincere, skillfully

communicative, smart and condescending one's partner was such that, in general, participants

receiving positive feedback perceived their partner to be more likeable, sincere, a skilled

communicator, smarter and less condescending than partners sending negative feedback. See

Table 3-6 for F and p-values, effect sizes, means and simple effects comparisons. Significant

main effects of feedback were also found on how liked participants felt by their partners, how

smart participants themselves felt, and how valid participants believed the test to be. Participants

receiving positive feedback thought they were liked more by their partners, thought the test was

more valid, and thought they were themselves smarter than participants receiving negative

feedback. Negative feedback participants believed their partner was more straightforward than

participants receiving positive feedback (p = .0015) or ambiguous feedback (p = .0008).

Ambiguous feedback partners were not seen as significantly different from positive feedback










partners on any of these measures. In short, partners who delivered positive feedback and

partners who delivered ambiguous feedback were seen positively and created positive feelings,

and both differed from partners who delivered negative feedback.

Hypothesis X. 1 and X.2 predicted that positive feedback partners would be most likeable

and would seem to like the participants themselves most while little difference would be found

between negative and ambiguous feedback partners. While positive feedback partners were

viewed more favorably than negative feedback partners, ambiguous feedback partners were

viewed just as favorably, thus failing to support the notion that ambiguous feedback partners

would suffer a tarnished reputation for their use of covert aggression. Hypothesis XI. 1, XI.2, and

XI.3 predicted that ambiguous feedback partners would be seen as the best communicators,

followed by positive feedback partners, and negative feedback partners would be seen as the

least effective communicators. Because positive and ambiguous feedback partners were viewed

as equally good communicators and significantly better communicators than negative feedback

partners, Hypothesis XI. 1 and XI.3 were both supported (that ambiguous feedback partners

would be seen as more skillful communicators than negative feedback partners and positive

feedback communicators may experience a halo effect and be seen as a better communicator than

negative feedback communicators). However, because ambiguous and positive feedback partners

were viewed similarly, Hypothesis XI.2 (that ambiguous feedback partners would also be seen as

more skillful communicators than positive feedback partners) was not supported. Therefore,

although technically Hypothesis XI received partial support in that both positive and ambiguous

feedback partners would be seen as better communicators than negative feedback

communicators, the main focus of this hypothesis predicted that ambiguous feedback

communicators would be seen as particularly cunning communicators in their use of covert










aggression, even relative to positive feedback communicators, and this was not found to be the

case.

Hypothesis XII predicted that participants receiving (1) negative and (2) ambiguous

feedback would perceive the creative intelligence test as less valid than positive feedback

participants. In actuality, ambiguous and positive feedback participants perceived the creative

intelligence test to be equally valid, and significantly more valid than negative feedback

participants, thus Hypothesis XII. 1 was supported but not Hypothesis XII.2. Hypothesis XIII

predicted that negative and ambiguous feedback participants would not internalize the

disparaging feedback they received and should not differ in how creatively intelligent they

believe themselves to be (although negative feedback participants would be more likely to

internalize than ambiguous feedback participants). Although negative feedback participants

reported that they believed themselves to be less creatively intelligent than ambiguous feedback

participants and positive feedback participants, this was not drastically different, and the

ambiguous feedback participants did not differ from the positive feedback participants in how

creatively intelligent they believed themselves to be. Thus, the spirit of this hypothesis was

generally supported.

An additional measure was included to assess whether participants varied regarding how

positively they believed they had evaluated their partner' s creative intelligence assessment prior

to completing the First Impressions Questionnaire. Participants did not differ in how positively

they believed they had evaluated their partner' s work as function of feedback (F (2, 219) = 0.04,

p = .96), response option (F (1, 219) = 0.99, p = .32), or a Feedback X Option interaction (F (2,

219) = 0.82, p = .44).









To assess whether perceptions of their partners were affected by whether participants

expected an interaction, a 2 (Interaction Expectation: Meeting or No Meeting) X 2 (Feedback:

Ambiguous or Negative) general linear model was conducted with the offset comparison groups

to assess whether meeting expectations predicted or moderated the above findings. With two

exceptions, interaction expectations did not predict or moderate how participants viewed their

partners. Participants who expected to meet their partner perceived the test to be somewhat less

valid than those who did not expect to meet their partner (F (1, 139) = 6. 12, p = .01, rl, 2 = .042,

Mmet = 5.15, M~No meet = 5.76). In addition, there was a marginally significant trend for

participants who expected a meeting to report that they had evaluated their partner more

positively (F (1, 139) = 3.24, p = .07, rl, 2 = .023, Mmeet = 7.49, M~ No meet = 7.09) suggesting that

perhaps participants who anticipated an interaction with someone wished to be a bit nicer than

those who did not expect to meet their partner.

Affect

To determine whether or not affect differed as a function of feedback, positive and

negative affect scores (Cronbach' s a = .87 and .86, respectively) were subj ected to a model with

feedback and response option as predictors. Positive and negative affect scores were slightly, but

significantly correlated, r (225) = -. 18, p = .005. Mean levels of affect by feedback are presented

in Table 3-6 along with F and p-values and effect sizes. A main effect of feedback was found for

both positive and negative affect. Negative feedback participants reported less positive affect and

more negative affect than positive feedback or ambiguous feedback participants (p < .0001); the

latter two conditions did not differ. Therefore, Hypothesis XIV. 1 and XIV.2 were supported (that

positive feedback participants would report positive emotions, and negative feedback

participants would report negative emotions) but Hypothesis XIV.3 (ambiguous feedback

participants would report both positive and negative emotions) was not supported.










To assess whether participants experienced differential affect as a result of expecting an

interaction, a 2 (Interaction Expectation: Meeting or No Meeting) X 2 (Feedback: Ambiguous or

Negative) general linear model was conducted with the offset comparison groups to assess

whether meeting expectations predicted or moderated the general above findings regarding

positive and negative affect. Interaction expectation did not predict either positive or negative

affect, alone or in combination with feedback.

Communication Preference

Participants were asked to select a method of communicating for their collaborative

interaction with their partner. To test the hypothesis that participants receiving negative or

ambiguous feedback would eschew face-to-face communication, communication method

preferences were submitted to a chi-square analysis with feedback. As shown in Table 3-7, the

results indicate a significant variation in communication method by feedback (X 2 = 19.85, p =

.0005), such that face-to-face communication was preferred by all feedback conditions, but

negative feedback participants preferred computer-mediated communication more than

ambiguous feedback participants or positive feedback participants. Therefore, Hypothesis XV

(that participants receiving negative and ambiguous feedback would be most likely to avoid face-

to-face interactions and instead seek out less immediate forms of communication such as

computer mediated interaction and intercom mediated interaction), found support in negative

feedback participants, but not among ambiguous feedback or positive feedback participants.

Moderators

A variety of individual difference variables were included as possible moderators of

effects. To determine the nature of any interactions, analyses were conducted using the SAS

General Linear Model (GLM) procedure on ranking scores. The predictors in the model were

feedback (3 levels: ambiguous, negative, or positive), option (2 levels: mixed message or neutral









message condition), each respective moderator (as a continuous, centered variable), and all

interactions. Moderators were centered so that any main effects of option or feedback could be

interpreted as an effect on a person who has an average score on the moderator. High and low

levels of each moderator were defined as 1 SD above and below the mean level; scores reported

for high, low, and average moderator reflect GLM estimates of the values of the dependent

measures at these integrity levels (see Cohen, Cohen, West & Aiken, 2003).

Perspective Taking

It was expected that participants higher in perspective taking would show an increased

likelihood of using covert aggression, i.e., endorsing mixed messages, due to previous research

findings that perspective takers were more capable of indirect aggression than low perspective

takers. Neither total empathy scores (a = .78) nor empathic concern scores (a = .76) produced

any significant affects on ranking of statements, alone or in combination with other predictors.

A three-way feedback X option X perspective taking interaction was found on ranking negative

messages, F (2, 208) = 4.81, p = .0091, r1,2 = .044, a perspective taking = .76. See Figure 3-1 for

estimated mean rankings generated by the SAS estimate command for participants high,

medium, and low in perspective taking as a function of feedback and option. Table 3-8 presents

these means along with comparisons across conditions and levels of perspective taking.

Closer examination of these interactions reveals that participants receiving positive or

ambiguous feedback were unaffected by either response option or perspective taking ability in

their selection or ranking of negative messages. In contrast, negative feedback participants

generally ranked negative messages higher in the neutral messages condition than in the mixed

messages condition and this tendency was exaggerated among participants high in perspective

taking. Hypothesis V predicted that participants higher in perspective taking would use more

mixed messages. This was not the case as perspective taking did not predict rankings of mixed









messages. However, perspective taking did predict rankings of negative messages in response to

negative feedback such that, in the mixed messages condition, high perspective takers did not

endorse negative messages as strongly as low perspective takers. On the other hand, in the

neutral messages condition, high perspective takers endorsed negative messages much more

strongly than low perspective takers. Perhaps high perspective takers in the mixed messages

condition perceived their endorsement of any mixed messages as sending negative messages and

so could back off the endorsement of negative messages. In contrast, because neutral messages

did not send this aggressive undertone, participants who wished to aggress were forced to more

strongly endorse negative messages in the neutral message condition.

Other Moderators

Several other possible moderators were tested including hostility (a = .78), anger (a = .82),

physical aggression (a = .85), direct aggression (a = .87), and indirect aggression (a = .83).

These moderators produced no significant effects either alone or in interaction with other

predictors. As such, the results failed to support Hypothesis IV (that participants with greater

proclivity toward direct aggression would show higher rates of overt aggression), and failed to

support Hypothesis III. 1, III.2, and III.4 (that participants with greater proclivity toward hostility,

anger, and indirect aggression would show higher rates of covert aggression). Verbal aggression

(a = .69), was found to significantly predict ranking of alternate messages (F (1, 208) = 4.61, p =

.03) such that less verbally aggressive participants endorsed alternate messages more than more

verbally aggressive participants, thus failing to support Hypothesis III.3 that more verbally

aggressive participants would show higher rates of covert aggression. Proclivity toward verbal

aggression did not predict rankings of negative messages.


3 Selection of alternative statements (as opposed to ranking) was only marginally
predicted by proclivity toward verbal aggression, F (1, 210) = 1.94, p = .17.









It was expected that individuals high in social desirability would be less likely to endorse

negative or mixed messages than positive messages, When these high social desirability

participants do act on their aggressive motives, though, they would endorse mixed messages

more than low social desirability participants, who would prefer outright negative messages over

mixed messages Social desirability significantly interacted with feedback and option in ranking

of negative messages (F (2, 208) = 4.33, p = .01, r12= .040), and ranking of positive messages (F;

(2, 208) = 3.01, p = .0513, r,2 = .028).4 See Figures 3-2 and 3-3 for mean rankings of negative

and positive messages. Table 3-9 presents these means along with comparisons across conditions

and levels of perspective taking. The interaction indicates that while negative feedback

participants' rankings were affected by social desirability and response option, positive and

ambiguous feedback participants generally were not. Specifically, negative feedback participants

in the mixed messages condition endorsed negative messages less than negative feedback

participants in the neutral messages condition. This general tendency was exaggerated among

high social desirability participants in the mixed messages condition such that high social

desirability participants actually endorsed negative messages significantly more than low social

desirability participants.

The inverse numerical pattern is true for rankings of positive messages, consistent with the

above pattern that high social desirability participants are most likely to engage in socially

desirable behavior when mixed messages were available rather than neutral messages as the

alternate message. In keeping with the general findings, negative feedback participants endorsed

4 The effects of social desirability and perspective taking reflect similar patterns on the
dependent measures. Social desirability was significantly correlated with perspective taking, r
(227) = 0.33. However, this correlation alone cannot account for the similarities as a similar
strength correlation was found between other moderators, e.g., social desirability was
significantly correlated with both verbal aggression and hostility (r (227) = -.31 and .45,
respectively, and these moderators did not significantly predict the dependent measures.









positive messages more in the mixed messages condition than the neutral messages condition,

but this was particularly true for those high in social desirability.

In sum, participants higher in social desirability pursued their socially desirable motives to

be nice in the face of negative feedback when mixed messages were available, but when only

neutral messages were available, high social desirability participants were significantly more

negative than their low social desirability counterparts. Therefore, Hypothesis VI. 1 and VI.2

were not supported--social desirability participants did not differ in their use of covert

aggression compared to low socially desirability participants. However, the presence or absence

of mixed messages as an option did affect participants' ranking of negative and positive

messages. In the wake of having mixed messages available, high social desirability participants

were considerably nicer than their low social desirability counterparts. In contrast, in the wake of

having neutral messages available, high social desirability participants were considerably more

aggressive than low social desirability participants.













Positive Interpretation 7.00a (1.70) 1.92 (1.26) 7.68a (1.25)
Negative interpretation 3.46 (1.53) 7.42 (1.45) 2.12 (0.88)
Double-meaning 5.35 (2.37) 2.85b, (2.34) 3.68b, (2.06)

Would be upset 3.27 (2.27) 5.15 (2.29) 2.04 (2.07)
Other trying to be superior 5.76, (2.22) 6.38, (2.23) 3.78 (2.15)
Other is concealing insult 4.92e (1.98) 4.23ef (2.41) 3.13f (1.66)
Other is sincere 5.38e (2.23) 5.96cd (2.07) 6.44d (1.53)
n 26 26 25
For each dependent measure, means sharing a subscript are not significantly different.


Table 3-1. Pilot study: Perceptions of feedback
Ambiguous
feedback


Negative
feedback


Positive
feedback









Table 3-2. Ranking means
Feedback Statement type
Ambiguous feedback
Alternative messages
Negative messages
Positive messages

Negative feedback
Alternative messages
Negative messages
Positive messages

Positive feedback
Alternative messages
Negative messages
Positive messages


Mixed messages


Neutral messages


9.18a
13.40cd
5.92,
45

8.41xi
10.78
9.31i
29

9.66ak
13.39gdl
5.45fn,
38


(1.07)
(0.48)
(1.22)


(1.72)
(2.40)
(3.05)


(1.40)
(0.38)
(1.43)


10.01b
13.47,e
5.01,
36

8.67hj]
9.163
10.66
35

9.64bk
13.44,1
5.42n,
37


(1.43)
(0.15)
(1.46)


(1.55)
(2.90)
(2.44)


(1.62)
(0.27)
(1.67)


Values reflect average ranking of messages. Therefore, lower values reflect more endorsement of
that category. Orthogonal simple effects comparisons are presented in the table, with means
having at least one factor in common and sharing a subscript not significantly different from one
another. Standard deviations are indicated in parentheses.










Table 3-3. Category ranking difference scores


Feedback Statement type
Ambiguous feedback
Alternate minus Negative
Negative minus Positive
Positive minus Altemnate
Negative feedback
Alternate minus Negative
Negative minus Positive
Positive minus Altemnate
Positive feedback
Alternate minus Negative
Negative minus Positive
Positive minus Altemnate


Mixed messages


Neutral messages


4.22
-7.48
3.26

2.37
-1.47
-0.90

3.73
-7.94
4.21


(1.12)
(1.51)
(2.24)

(2.85)
(5.22)
(4.33)

(1.47)
(1.56)
(2.81)


3.46
-8.46
5.00

0.49
1.50
-1.99

3.79
8.02
4.23


(1.42)
(1.50)
(2.89)

(3.95)
(5.13)
(2.89)

(1.61)
(1.77)
(3.28)


Values represent differences in category rankings. In the first cell, 4.22 indicates that
participants receiving ambiguous feedback in the mixed messages condition rated mixed
messages 4.22 ranks above negative messages, on average. Standard deviations are indicated in
parentheses.









Table 3-4. Ranking means with offset comparison groups not expecting a future interaction
Feedback Statement type Future meeting No future meeting
Ambiguous feedback
Mixed messages 9.18a (1.07) 9.28a (1.21)
Negative messages 13.40b (0.48) 13.39b (0.32)
Positive messages 5.92, (1.22) 5.84, (1.11)
n 45 36
Negative feedback
Mixed messages 8.41de (1.72) 7.80d (1.97)
Negative messages 10.78f (2.40) 10.40f, (2.06)
Positive messages 9.31e (3.05) 10.30, (2.08)
n 29 31
Values reflect average ranking of messages. Therefore, lower values reflect more endorsement of
that category. Orthogonal simple effects comparisons are presented in the table, with means
having at least one factor in common and sharing a subscript not significantly different from one
another. Standard deviations are indicated in parentheses.









Table 3-5. Trait nroupinns and relevant Cronbach alphas
Trait a
Liking of the other .91
Do you believe that you will like the other participant?
Do you think that the other participant is generally friendly?
Do you believe you will respect the other participant?
How much do you think you will enjoy working with the other participant?
How well do you think this person gets along with other people?
How much are you worried about interacting with the other person?*
Intelligence of the other .82
How much creative intelligence do you think the other participant possesses?
How much do you value the other participant' s opinion regarding creative intelligence problem solving?
Evaluation by the other .95
How much creative intelligence does the other participant believe you have?
How positively did the other person evaluate your work?
Straightforwardness of the other
To what extent do you think that the other person is straightforward?
Sincerity of the other
To what extent do you think that the other person is sincere?
Communication skill of the other
To what extent do you think that the other person is a skilled communicator?
Liking by the other .91
Do you believe that the other participant will like you?
How much do you think the other participant respects you?
Test validity .88
How well do you think the first assessment of creative intelligence actually assessed creative intelligence?
How well do you think the second assessment of creative intelligence will assess your creative intelligence?
Self intelligence
How much creative intelligence do you believe you possess?
Evaluation of other' s individual creative intelligence assessment
How positively did you evaluate the other person's work?
*indicates item was reverse scored










Table 3-6. Perceptions of partner and affect
Dependent measure Ambiguous Negative Positive Error
feedback feedback feedback df F p ],2


96a (0.87)
33a (1.13)
58a (1.21)
83a (1.43)
78a (1.26)
20a (1.33)


4.02 (1.18)
4.28 (1.62)
8.42 (0.72)
7.55 (1.25)
4.64 (2.11)
4.70 (1.95)


7.16a (0.58)
7.15a (0.89)
2.34a (0.68)
6.87a (1.02)
6.94a (1.05)
7.17a (0.95)


221
219
219
221
221
221

221
219
219
219


265.85
131.44
943.86
7.06
50.07
68.41

219.25
54.63
8.79
0.04


<.0001
<.0001
<.0001
.0011
<.0001
<.0001

<.0001
<.0001
.0002
.9566


.706
.546
.896
.060
.312
.382

.662
.333
.074
.000


Liking of the other
Intelligence of the other
Evaluation by the other
Straightforwardness of the other
Sincerity of the other
Communication skill of the
other
Liking by the other
Test validity
Self intelligence
Evaluation of other' s assessment


6.'
7.:
2.1
6.'

7.:


6.68a
6.43a
6.70a
7.36a


(0.98)
(1.47)
(1.15)
(1.17)


3.37
4.38
5.80
7.32a


(1.27)
(1.71)
(1.68)
(1.41)


6.67a
6.69a
6.58a
7.38a


(0.99)
(1.04)
(1.32)
(1.10)


Positive affect 27.94a (6.89) 23.03 (7.75) 29.21a (7.32) 220 13.44 <.0001 .109
Negative affect 13.46a (5.36) 18.23 (6.41) 12.58a (3.07) 220 24.48 <.0001 .182
For each trait or affect, means that share a subscript are not significantly different. Subscripts a and c denote marginally significant
differences, p = .15 and p = .11, respectively. Standard deviations are indicated in parentheses.













Ambiguous 80 0 3 83
(96.39%) (0.00%) (3.61%)
Negative 47 2 15 64
(73 .44%) (3.13%) (23.44%)
Positive 67 1 5 73
(91.78%) (1.37%) (6.85%)
Values not in parentheses indicate number of individuals choosing each communication method.
Values in parentheses indicate percentage of participants in each feedback condition selecting
that particular communication method.


Table 3-7. Participant communication method preferences
Chosen communication method
Feedback Face-to-face Spae yt Cmur


Total










Table 3-8. Ranking of negative messages as a function of perspective taking, feedback, and response option
Response
Mixed Neutral option
Feedback messages messages comparison
Ambiguous feedback
Perspective slope: ns ns
High 13.36 13.49 ns
Average 13.42 13.47 ns
Low 13.47 13.46 ns
Negative feedback
Perspective slope: **
High 11.34 8.01 ****
Average 10.79 8.82 ****
Low 10.25 9.63 ns
Positive feedback
Perspective slope: ns ns
High 13.45 13.46 ns
Average 13.36 13.44 ns
Low 13.28 13.42 ns
For each level of feedback and perspective taking, the response option comparison is indicated in the far right column ****" indicates p
<.0001, **" indicates p < .01, ns indicates p > .05. The significance of perspective taking for each combination of response option and
feedback is indicated each set of means. Negative feedback differed significantly from both positive and ambiguous feedback for both
response options at all levels of perspective taking; positive feedback did not differ significantly from ambiguous feedback at any
level of perspective taking for either response option.










Table 3-9. Ranking of positive and negative messages as a function of social desirability,
feedback, and response option
Feedback Response
Mixed Neutral option
messages messages comparison
Ambiguous feedback
Negative ranking
Social des. slope ns ns
High 13.48 13.48 ns
Average 13.40 13.47 ns
Low 13.32 13.46 ns
Positive ranking
Social des. slope ns ns
High 5.83 5.12 ns
Average 5.92 4.98 *
Low 6.01 4.84 .06
Negative feedback
Negative ranking
Social des. slope ** *
High 11.84 8.56 **
Average 10.95 9.13 ****
Low 10.06 9.70 ns
Positive ranking
Social des. slope ns
High 8.27 10.98**
Average 9.14 10.68 ***
Low 10.01 10.38 ns
Positive feedback
Negative ranking
Social des. slope ns ns
High 13.42 13.46 ns
Average 13.40 13.44 ns
Low 13.37 13.41 ns
Positive ranking
Social des. slope ns ns
High 5.65 5.05 ns
Average 5.47 5.45 ns
Low 5.28 5.85 ns
Within each level of feedback and social desirability, the response option comparison is
indicated between the columns of means. ****"indicates p <.0001, ***indicates p < .001,
*indicates p <.05, ns indicates p > .05. The significance of social desirability for each
combination of response option and feedback is indicated above each set of means. For all three
dependent measures (negative ranking and positive ranking) negative feedback differed
significantly from both positive and ambiguous feedback for both response options at all levels
of social desirability; positive feedback did not differ significantly from ambiguous feedback at
any level of social desirability for either response option.

















Negative Message Ranking


SAmbiguous Feedback
Mixed Messages


Ambiguous Feedback
Neutral Messages


- Negative Feedback
Mixed Messages


* A Negative Feedback
Neutral Messages

*********** ******~~ Positive Feed back

Mixed Messages

******* **A** ******~ Positi ve Feedbac k

Neutral Messages


*
"
, .*** "


r
r'
,r
r


Lz
z


I
I
I
I


A- .


I
I
Z
r
r


Average

Perspective Taking


High


Figure 3-1. Mean ranking for negative messages as a function of perspective taking, feedback, and response option. Lower values

indicate more endorsement of negative messages.





















A m e m iuli lnlll1111111 111111111uili liluilslilill aislisluilui= ===== = ====== = = ** = = -


I I


Negative Message Ranking


SAmbigous Feedback
Mixed Messages


Ambiguous Feedback
Neutral Messages

* *Negative Feedback
Mixed Messages

* A *Negative Feedback
Neutral Messages

--<>*--~~~~~~~~Positive Feedback
Mixed Messages

""~~~~+~~~~ f"Positive Feedback
Neutral Messages


12






a

a 10


. g


*
* I


. o *


L
I
I


L
L
I
I
I


am *


& .** ..


" .u .

I A


Low


Average

Social Desirabilit


High


Figure 3-2. Mean ranking for negative messages as a function of social desirability, feedback, and response option. Lower values
indicate more endorsement of negative messages.














































A.........,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 1


1


Positive Message Ranking


e,,r'"'L

r
I
I
Z
r


SAmbiguous Feedback
Mixed Messages


Ambiguous Feedback
Neutral Messages


* Negative Feedback
Mixed Message


* A Negative Feedback
Neutral Messages


*****************~~~Positive Feedback
Mix ed Messages


*****A******~~~~~Positive Feedback
Neutral Messages


* .g


I
I
I
Z
r
r
r


I
I
'


Low


Average
Social Desirability


High


Figure 3-3. Mean ranking for positive messages as a function of social desirability, feedback, and response option. Lower values

indicate more endorsement of positive messages.









CHAPTER 4
DISCUSSION

It was predicted that participants receiving ambiguous feedback conveying a mixed

message would interpret this feedback as covert aggression, and these participants would prefer

to respond to such a message with covert aggression themselves by endorsing mixed message

statements. In contrast, it was predicted that recipients of ambiguous feedback who were not able

to respond with their own mixed message statements, would be cautious and resort to primarily

positive messages. However, participants did not react to ambiguous feedback as expected and

instead responded to their partners in a fashion quite similar to the manner in which positive

feedback participants responded to their partner--in a primarily positive manner.

The availability of a mixed message option did produce some subtle effects on ambiguous

feedback and negative feedback recipients. Specifically, while ambiguous feedback participants

did not endorse mixed messages very strongly, they did endorse mixed messages more strongly

than neutral messages (with a compensating weakened endorsement of positive messages),

indicating some sensitivity to the message option manipulation and the subtle differences

between neutral and mixed messages. One possible interpretation of this pattern is that

ambiguous feedback recipients were indeed sensitive to the duality of both the feedback they

received and the messages they sent. If so, this pattern of message endorsement indicates that

these participants did perceive the feedback as a possible slight and were responding, subtly, via

slightly stronger endorsement of mixed messages. However, this possibility is not consistent with

other measures. Ambiguous feedback participants felt similar levels of positive and negative

affect as positive feedback participants, indicating that they were not more upset by the partner's

assessment of their work than positive feedback participants. Moreover, ambiguous feedback










participants reported perceiving their partner as positively as positive feedback participants saw

their partner.

Although it is possible that ambiguous feedback participants did perceive the duality in

their feedback and employed the negative interpretation of the mixed messages in retaliation, this

is unlikely. Instead, what is more consistent with the other reports from the ambiguous feedback

participants is that they were as motivated as positive feedback participants to send positive

messages, and endorsed mixed messages because of their positive interpretation while oblivious

to the negative interpretation. In short, their attempt to be positive was diverted between positive

messages and mixed messages. The same attempt to send positive messages would not cause a

strong endorsement of neutral messages because neutral messages were clearly less positive than

positive messages and the positive aspect of mixed messages. This interpretation is consistent

with the finding that positive feedback and ambiguous feedback participants did not significantly

differ from one another in their use of positive messages, negative messages, or mixed messages.

The only real difference between positive feedback and ambiguous feedback participants was

that ambiguous feedback participants were somewhat affected by the mixed message option

whereas positive feedback participants were not.

In contrast, while negative feedback participants endorsed mixed messages and neutral

messages equally, there are indications that negative feedback participants may have still been

sensitive to the dual nature of the mixed messages. More specifically, after negative feedback,

participants were more negative and less positive in their messages if neutral message options

were available than if mixed message options were available. On the surface, then, those who

had mixed message options seemed friendlier. Yet, regardless of the message option condition,

those who received negative feedback experienced negative affect, rated the other person










negatively, and clearly disliked that other. This suggests that the availability of the mixed

messages may have led participants to become less hostile, which seems unlikely given their

later negative affect and ratings of the other, or that they used the mixed messages as "digs." The

latter interpretation is consistent with the ratings of the pilot participants and the intent of the

manipulation. If this latter interpretation is correct, then these participants may have then felt

endorsing positive messages better disguised their true motivation for endorsing mixed

messages. Moreover, after receiving negative feedback, these participants may have been more

attune to the negative connotations of mixed messages whereas positive and ambiguous feedback

participants were not as "tuned in" to the possible negative connotations of the mixed messages.

Thus, ambiguous and positive feedback participants may have been less aware of the negative

connotations of the negative messages.

Although negative and ambiguous feedback participants were subtly influenced by the

message option manipulation, positive feedback participants were not at all affected by this

manipulation. Participants receiving positive feedback did not differ in their use of positive

messages, negative messages or alternate messages as a function of message option. Instead,

positive feedback participants chose to be primarily positive in their messages sent to their

partner, avoided overtly negative messages altogether, and did not treat mixed messages and

neutral messages differently, ranking these two types of messages similarly to each other; these

messages were ranked intermediate between negative messages and positive messages.

Comparisons between Positive and Ambiguous Feedback

It may seem, on the face of it, that participants did not perceive much difference between

the positive feedback and ambiguous feedback. However, pilot testing clearly indicated that

ambiguous feedback was seen as significantly more upsetting, more negative, as conveying a

negative meaning at the same time as a positive meaning, as concealing more of an insult, being









less sincere, and as more of an attempt to be superior. If anything, pilot testing indicated that

participants should have perceived the ambiguous feedback as much more like the negative

feedback than the positive feedback.

There are at least two possible interpretations of this pattern of relatively few differences

between positive feedback and ambiguous feedback in the face of pilot evidence that indicates

there should be many differences between these two feedback conditions. One possibility is that

although participants clearly identified the multiple nuanced interpretations of the feedback, they

preferred to "play it safe." In other words, although the ambiguous, covertly aggressive feedback

was upsetting, in deciding how to respond, participants considered the consequences of

responding aggressively (covertly or overtly) and being wrong as more dire than responding

nicely and being wrong. In keeping with this line of reasoning, ambiguous feedback participants

may have identified the possible dual interpretations, but may have felt foolish to act on

something that may have been estimated to be a very unlikely possibility, and may have then

reported positive estimations of their partner.

If this is the case, then a couple of situational factors may have caused participants to

decide it a better choice to play it safe. First, the negative interpretation of the ambiguous

feedback may not have been seen as negative enough to risk the consequences of retaliating. In

other words, although the negative interpretation may have been identified, the participants may

not have seen this negative connotation as so strong that it deserved a response. A second

possible factor is that instead of seeing the upcoming interaction and creative intelligence

assessment as so important that negative communications should not be risked, is that perhaps

the participants actually perceived little value in their performance on the upcoming assessment

and so were not motivated to defend themselves as much as they would if the assessment were










highly important. Although this second option seems possible given the sometimes apathetic

nature of participants in general, this does not seem to be the most parsimonious explanation as

some participants were visibly upset by the feedback they received and clearly took it personally

(as reported anecdotally by research assistants serving as experimenters). Moreover, participants

reported the measure to be a generally valid assessment of creative intelligence--even the

negative feedback participants (who saw the test as least valid) on average reported that they

perceived the test to be a moderately valid measure of creative intelligence with scores near the

midpoint of a 9-point scale.

Another possibility for the general lack of differences between the positive feedback and

ambiguous feedback conditions can be drawn from the literature on ingratiation. This literature

indicates that recipients of ingratiation attempts are generally quite willing to believe that the

ingratiation directed at them is sincere. However, witnesses to ingratiation attempts who are not

themselves the target of the ingratiation attempt can readily identify the ingratiator' s true motive

(Jones, 1964; Vonk, 2002). This distinction between the person on the receiving end of an

insincere compliment and an observer of such an act may elucidate the general failure to find

differences between responses to positive and ambiguous feedback. Specifically, in the current

study, participants receiving ambiguous feedback may have reacted the same way that a target of

ingratiation responds: accept the insincere compliment at face value. In contrast, the participants

in the pilot study evaluated the interpretations of the feedback knowing that the feedback was not

from a real person in response to their actual work. Therefore, personal biases that come from

being the true recipient of an insincere compliment may not have come into play.

People who find themselves as the target of ingratiation may not interpret insincere

compliments accurately for several reasons. Most obviously, participants may be motivated to









interpret it as sincere (Jones, 1964). Not only does this provide a boost to one' s self-worth

(Vonk, 2002), but it is also simpler to interact with a person who is sincere rather than insincere,

and people may be motivated to avoid believing that they interact with insincere others.

Additionally, Vonk (2002) proposed that people actively engaged in interaction with another

person would have fewer cognitive resources available to second-guess a person's motives than

someone who is not actively engaged in conversation and having to monitor, plan, and evaluate

their own contribution to the interaction. In short, witnesses are able to focus more attention on

the potential insincerity of an ingratiator/covert aggressor and thus more likely to detect

insincerity. Although his sole test of this possibility did not receive strong support, Vonk

assessed situations in which the consequences for insincerity are not as dire (i.e., a potential

criticism and an upcoming evaluative interaction), as in the current study. In contrast, with the

heightened consequences of the communication in the current study, the cognitive business

explanation may still be a viable explanation for the current data.

Alternatively, Vonk (2002) also assessed the possibility that a heightened ingratiation--

self-concept discrepancy may result in a further consideration and questioning of the ingratiator' s

motives and sincerity. Vonk assessed this possibility via self-esteem and found that low self-

esteem recipients of flattery believed ingratiators' comments to be less accurate than high self-

esteem flattery recipients. Vonk concluded that these low self-esteem participants experienced a

greater discrepancy between their self-view and the feedback they received from the ingratiators

and were therefore more motivated to seriously evaluate the sincerity of the feedback. In

contrast, the high self-esteem participants did not experience the same self-view--feedback

discrepancy and so they were not motivated to reassess the sincerity of the feedback.









This alternative vantage point explanation is in some ways illustrated via the discrepancy

in how ambiguous feedback participants viewed the dual nature of the feedback they received

and the messages they sent to their partner. If these participants did not see the dual nature of the

mixed messages, they would not have avoided these statements. However, participants who

presumably saw little insincerity in the evaluation of their own work significantly preferred

positive messages to mixed messages. In other words, while these participants were oblivious to

the insincerity of the messages they received, they were at least somewhat aware of the potential

insincerity of the mixed messages they considered sending. Otherwise, these messages should

have been endorsed as often as the positive messages--they were not.

One possibility for future research would be to examine the extent to which witnesses to

such covert aggression alter their interaction behavior toward the victim or the perpetrator of the

aggression. Even if the covert aggression goes unnoticed by the intended victim, the aggression

may still have consequences for the victim or the aggressor. For example, as previous research

has suggested (LaFontana & Cillessen, 2002; Salmivalli, Kaukiainen, & Lagerspetz, 2000),

indirect and relational aggression are predictive of peer acceptance and social status. This

increase in social status should parallel a similar increase in respect (i.e., just as social status is

boosted as a results of aggression, respect should also be increased after the use of aggression).

Previous research does not specify whether this boon in social status is echoed in the victim of

the aggression. After any given instance of aggression, a person's liking for their aggressor is

almost certainly diminished, but the victim may also come to respect the aggressor more.

However, given the dual nature of covert aggression and the requisite savoir faire necessary to

properly encode the dual meaning, it is quite possible that individuals choosing to engage in

covert aggression recognize that the targets of such aggression may never realize that they have










been insulted. At times, it may even be plausible that covert aggression is intended solely for the

purpose of impressing witnesses to such events. As such, covert aggressors' goal would not be

simply to cause psychological pain in their victim, but rather the purpose is much more

personal--to impress witnesses (who should be more capable of determining the insincerity of

the covert aggression) with the aggressor' s own intellect, communication skill, and willingness

to attack another.

The awareness of the social ramifications of covert aggression does not preclude that

occasionally, people may engage in covert aggression without witnesses to the event or with the

assumption that their target will not understand the intended slight. In fact, they may find the

gullibility of the victim all that more entertaining. In short, people may engage in covert

aggression simply for their own personal pleasure. However, to the extent that covert aggression

is intricately linked to the presence of third parties, then future research should examine how and

to what extent the presence of a third party in the same social fiber affects the rate of covert

aggression and the effectiveness of it. The presence of others should increase the use of covert

aggression over overt aggression not only because it portrays the aggressor as a sophisticated

social creature, but also because the presence of others within the same social network amplifies

concerns with maintaining a cool, sophisticated, non-brutish image. In contrast, if covert

aggression is truly intended as a social tool, then the presence of a third party not relevant to

ongoing social interactions should have little effect on the implantation of covert aggression.

Nature of the Interpersonal Relationship

In addition to the possibility that covert aggression is primarily a social tool, future

research should investigate and be able to demonstrate that changing the nature of the social

situation should similarly affect use of covert aggression. For example, covert aggression should

be a particularly successful tool for those who Eind themselves in a situation that prohibits overt










aggression--such as aggressing against a higher-status victim. As such, the covert to overt

aggression ratio should be heightened as the appropriateness of engaging in overt aggression

diminishes. For example, some cultures (e.g., southern culture) discourage open aggression

(Cohen, Vandello, Puente, & Rantilla, 1999) and as a result, covert aggression should be even

more likely. The existence these kinds of norms do not necessarily have to reflect larger cultural

expectations, however. Such variations could also be a reflection of a specific peer-group's

expectations for social behavior. Just as some individuals would rather hear a criticism in a

straightforward manner and others prefer sugarcoated criticism, there is also variability in the

degree to which certain friendship groups prefer open and honest communication versus

aggression disguised as something sweet.

Other factors that may affect the use of covert aggression include not only the relative

status of individuals or the norms for interactions, but also perceived security in a particular

relationship. Covert aggression, as a social tool, is presumably more important in some

relationships than others. Specifically, relationships vary in the degree to which they can be

characterized by the game-playing nature of covert aggression. Some relationships, such as the

early stages of romantic relationships are much more characterized by game playing, in that they

involve a strategic approach and reaction to the other person's behavior. However, as

relationships grow and mature, the role that such power concerns play decreases. The

characterization of a mature relationship marked by openness and concern for the other does not

have room for covert aggression. In fact, the use of covert aggression in such relationships may

have quite the adverse effect on the relationship partner and any witnesses to such behaviors.

Engaging in covert aggression in truly intimate relationships is likely not only to be more









difficult given that an intimate partner will be less easily duped, but the expectations for many

relationships is to discuss concerns openly. Doing otherwise may indicate a mark of immaturity.

Gender

Another noteworthy pattern is the lack of impact that gender of participants made. A great

deal of literature and a number of authors would argue that gender plays an important role in

who aggresses and how. Popular social psychological opinions hold that females aggress

verbally and indirectly whilst males aggress directly and physically (e.g., Galen & Underwood,

1997). If so, females in this study would have responded more negatively than men after

receiving negative feedback, and perhaps would also have been particularly reactive to instances

of covert aggression and the use of covert aggression. However, this was not found to be the

case. In fact, there were essentially no gender differences in use of overt or covert aggression, or

in response to overt aggression, covert aggression, or no aggression (positive messages).

Although this may seem surprising, one interpretation of these findings would actually be

consistent with previous research regarding social network density (Green, Richardson, & Lago,

1996). This research suggests that proclivity to engage in physical or direct aggression versus

verbal, relational, social, or indirect aggression, does not result necessarily from gender itself, but

rather should be considered under resources available to men and women. Specifically, women

tend to find themselves more heavily enmeshed in complicated social networks; their networks

are generally larger and individuals in these denser social networks tend to know other members

in the woman's social network. In other words, in women's network, there tend to be many

social connections shared among members of the social network. In contrast, men tend to have

smaller social networks and these networks tend to be more fragmented with each individual in a

man' s social network being considerably less likely to know other members of that social

network than in a woman' s social network. This greater social connectedness makes engaging in









indirect/verbal/social/ relational/covert aggression not only easier for women than men, but also

more effieacious. Men who choose to aggress using this strategy not only have fewer resources

with which to carryout the aggression (e.g., spreading a rumor is harder when the rumor can only

be spread to one person), but this aggression also has less social impact (the resulting negative

evaluation of the victim due to the deleterious rumor will harm the victim less when only one

person knows the rumor rather than many). Therefore, choosing to engage in nonphysical

aggression is a much wiser choice for female aggressors than male aggressors. Moreover,

engaging in this kind of aggression may spare a great deal of resources (physical energy,

potential bodily harm, etc.), over physical aggression.

This disparity in the efficiency of covert aggression as a function of social network density

is one factor that leads men to avoid this kind of aggression. However, Green, Richardson, and

Lago (1996) also demonstrated that men who have higher social network density demonstrate an

increased proclivity to engage in indirect aggression (and thus possibly covert aggression). In

other words, when the social environment of men and women are equated, gender differences in

indirect aggression are muted. In the current study, male and female participants found

themselves in an identical social situation: they would soon be temporarily participating with one

other same-sex person they did not know. In effect, this held the social environments of men and

women equal. Following from the results of Green et al. (1996), this provides men and women

the same social network resources and should therefore eliminate gender differences. As such,

the lack of gender effects is perfectly consistent with prior research.

Of course, other possible explanations for the lack of gender differences could be a simple

statistical anomaly. However given the prevalence of null findings involving gender, this

possibility seems, at best, a weak argument. Alternatively, perhaps cultural influences on










available aggressive methods have altered the manner in which males and females engage in

aggression. Specifically, as this age group becomes increasingly more reliant on non-face-to-face

communication as a way to maintain interpersonal ties (i.e., cell phones, instant messaging,

online social network communities, etc.), perhaps individuals--males and females alike--have,

as a generation, begun to shift away from more direct forms of aggression. With less reliance on

face-to-face interaction, there is perhaps less opportunity for males to practice direct or physical

aggression. As a result, they may develop similar aggressive behaviors as females.

Expectation of a Future Interaction

Another direction that future research should explore is how a long-term relationship

affects the nature of covert aggression. The current study attempted to investigate covert

aggression within the context of a somewhat dependent relationship that participants could not

readily avoid. Participants receiving negative feedback who did not anticipate a future interaction

did endorse mixed messages more strongly than negative feedback participants not expecting an

interaction, and also gave a weaker endorsement of positive messages when they did not expect

an interaction. However, the future interaction manipulation did not have the hypothesized

effect. It was predicted that participants facing an interaction would strongly endorse mixed

messages, but participants not facing an interaction would feel less social concern about the

consequences of engaging in overt aggression and would therefore abandon mixed messages for

straightforward negative messages. However, not only were ambiguous feedback participants

unaffected by the future meeting manipulation, but negative feedback participants were actually

more likely to use mixed messages when they did not anticipate a future interaction then when

they were.

One possibility for this pattern supports the above conjecture that participants were

generally being extremely cautious in their responses to ambiguous feedback. Specifically, an









imminent interaction should provoke participants to be more cautious than participants not facing

an imminent interaction. Therefore, in the face of an obviously negative attack, negative

feedback participants expecting an interaction were less cautious and more strongly endorsed

negative messages than ambiguous feedback participants expecting an interaction. However,

negative feedback participants not expecting an interaction were not as motivated to be cautious

and could therefore afford to be more negative, resulting in a stronger endorsement of mixed

messages. This finding is not inconsistent with the reasoning presented in this paper that

participants not anticipating a future interaction will be less concerned about the consequences of

their behavior than those expecting an interaction. This interpretation does of course assume that

negative feedback participants were aggressively motivated and perceived mixed messages as

conveying a negative interpretation. Pilot testing clearly indicates that these messages were

indeed perceived as having dual meanings, and the general eschewing of these messages by

positive and ambiguous participants indicates that the duality of meanings did not go entirely

unnoticed. Of course, these negative feedback participants not expecting an interaction were not

recklessly abandoning all thoughts of social consequences and caution: they did not strongly

endorse outright negative messages any more than negative feedback participants expecting an

interaction.

Perhaps, despite instructions that some participants would not work on the second creative

intelligence assessment with their partner, participants believed that they would in fact be

meeting their partner at the end of the session or for some other task. Indeed, although the

manipulation was generally successful, it was not a perfect manipulation and some participants

who were told they would not be completing the second assessment with their partner still

reported that they would be meeting their partner. This could have weakened the effect of this










manipulation, making participants who should not have expected a meeting to be more cautious

in the messages they sent.

Another possibility is that the nature of the future interaction did not carry the

psychological weight that is necessary to engage real life patterns of covert aggression. Perhaps

the prospect of meeting another undergraduate with whom one would interact only briefly was

not sufficient motivation for participants to be concerned with the other's true intentions or worth

stooping to aggression (covert or overt). Future research should aim to increase the psychological

magnitude of the future meeting, participants' desire to maintain positive interpersonal

relationships, and the consequences of not aggressing (e.g., lowered sociometric status,

decreased respect within a group, etc.). Ideally, such research would include investigation of

ongoing interpersonal groups. Unfortunately, the logistics of such a proj ect make this a difficult

task to achieve. For example, previous unpublished research (Miller & Schlenker, 2008),

attempted to study covert aggression in actual friendship pairs. In order to study these friendship

pairs, participants were asked to bring a friend to the lab. The resulting attempts to provoke

participants to covertly aggress against their friends were unsuccessful and essentially backfired

when participants with the most motivation to aggress against their friends being the least likely

to do so. Norms for appropriate friendship behavior were apparently heightened when the

opportunity to break those norms presented themselves. Moreover, friends who agree to

participate in a psychology study are likely close enough that it would take very extraordinary

circumstances to provoke true aggression against each other--circumstances that are not readily

reproduced in a psychological laboratory.

Perhaps experimental psychological research would be most successful in demonstrating

covert aggression between mere acquaintances. In such situations, motivation to maintain a non-


100










aggressive image might be particularly high. Another alternative methodology to investigate the

nature of covert aggression would be to create real experimental groups who would meet as part

of an ongoing longitudinal study. As such, participants would develop ties to the other

participants, be concerned with their image in the group, and be accountable to the group for

their behavior.

Social Desirability

It was predicted that participants higher in social desirability would be less likely to

endorse either negative message or mixed messages than low social desirability participants, but

if they did, they would show a preference for mixed messages over negative messages. In the

wake of negative feedback, participants with mixed messages available behaved quite in line

with the traditional nature of social desirability and this hypothesis: higher social desirability

participants were less negative and more positive in their messages than low social desirability

participants. However, when participants could only send neutral messages instead of mixed

messages, negative feedback participants responded quite differently. When high social

desirability participants could not send mixed messages, they were significantly more negative

than their low social desirability counterparts. In other words, when their hands were figuratively

tied and they were forced to be direct with their negative messages, high social desirability

participants exhibited more aggression.

In some ways, this may seem counterintuitive. These high social desirability participants

are most concerned with maintaining a positive interpersonal image, and strongly endorsing

negative messages does not seem to be the most effective method of achieving this goal. A

couple of possibilities exist that may explain this pattern. One possibility is that these

participants were more bothered by the negative feedback than low social desirability

participants. As a result, the high social desirability participants perceived the feedback as more









deserving of a negative response than low social desirability participants. However, although this

possibility seems plausible, participants high versus low in social desirability did not differ in

their emotional responses or views of the partner as a function of feedback.

An alternative account suggests that perhaps individuals high in social desirability

regularly avoid straightforward negative messages by way of creating in their daily interactions

ways to convey their own mixed messages. Left to their own devices, then, high social

desirability participants may seemingly exhibit a lower rate of obvious aggression while they are

in fact finding their own way to still be aggressive. Research on catharsis theory would suggest

that by engaging in aggression (albeit socially presentable aggression), people will be more

likely to aggress in the future. Most of the time, this increased aggression in high social

desirability individuals will still be expressed in a more covert manner. However, when high

social desirability individuals do not have the luxury of creating a socially acceptable way to

express their aggression, they may actually exhibit higher rates of open aggression than those

individuals who are not so creative in their aggressive endeavors.

Perspective Taking

Based on previous research regarding the role of perspective taking on successful

aggression, it was expected that participants highest in perspective taking ability would be the

most likely to engage in covert aggression by strongly endorsing mixed messages. However, this

was not the case as perspective taking did not predict rankings of mixed messages or neutral

messages. While this finding does not support the original hypotheses regarding use of mixed

messages, perspective taking wa~s predictive of negative message endorsement. In keeping with

the general pattern, participants receiving negative feedback endorsed negative messages more

strongly when neutral messages were available than when mixed messages were available.

However, this preference for neutral messages over mixed messages is particularly evident in










high perspective takers. In essence, high perspective takers were indeed more aggressive via

negative messages when they did not have the ability to engage in mixed messages. In contrast,

when mixed messages were available, high perspective takers were actually less aggressive than

their low perspective taking counterparts.

One possible interpretation of this is that high perspective takers may exhibit more overt

aggression but when other options are available, high perspective takers are not more aggressive.

Given that negative messages in this study parallel closely indirect aggression (not face-to-face

aggression that does not possess dual meanings), and traditional indirect aggression literature has

not allowed for the possibility of mixed messages, this is consistent with previous research

finding that high perspective takers can be more indirectly aggressive. Much of the existing

literature regarding indirect aggression and perspective taking has focused on younger children

and teenagers. If covert aggression is a more crafted and honed form of aggression, then perhaps

children and teenagers essentially do not have covert aggression available to them--not because

of situational constraints, but rather because they have not learned how to successfully employ

such a tactic. In contrast, more socially sophisticated young adults may have learned the art of

employing mixed messages in their aggression and favor this over more straightforward forms of

aggression, and as such, perspective taking leads to an inverse relationship in straightforward

aggression.

Distinguishing Covert Aggression from Other Forms of Aggression

One question that can be raised regarding these findings is the extent to which covert

aggression differs from indirect aggression or, more broadly, aggression in general. Covert

aggression shares many features with indirect aggression in the sense that both have an element

of concealment to them. Indirect aggression does not try to disguise the aggression, but rather

simply eliminates the opportunity for victims to immediately defend themselves. In contrast,










covert aggression may be done face-to-face, and while the victims of covert aggression have the

opportunity to aggress in retaliation if they so choose, the victims must first discern whether the

incident was actually malevolent. As a result, victims may postpone any immediate response not

because of inability (as is the case with indirect aggression), but because further consideration

may be warranted.

This similarity may result in future research ignoring the distinction between the two.

However, this could be a mistake. Covert aggression is a unique construct that introduces

different interpersonal costs and benefits for those choosing to engage in it. For example, if

asked to defend one' s aggressive behavior, an indirect aggressor' s only defense is a lack of

witnesses, which is something not always available to the indirect aggressor--simply because

the victim was not present for the behavior does not mean that others were not present. In

contrast, the covert aggressor's defense lies in whether the behavior can truly be considered

aggression or not. The covert aggressor would not deny the behavior occurred, but rather deny

the aggressive interpretation was intended. In other words, there is no perfectly decisive indicator

that the covert aggressor intended aggression. Future research should address the different

interpersonal consequences of engaging in covert aggression. At any single moment in time,

victims or witnesses may give the covert aggressor the benefit of the doubt and apply a

nonaggressive attribution to such behavior, but the kernel of doubt will be magnified with

repeated exposure to a covert aggressor' s shifty explanations for his/her behavior and will

eventually come to view this person as an aggressive individual, despite no single aggressive act

having enough evidence to support that conclusion. In many ways, this is similar to the

effectiveness of excuses: although excuses for poor performance do indeed work in the moment

to avert negative attributions about skill level, they negatively affect the holistic assessment of a


104










person's reliability and trustworthiness. Similarly, covert aggressors will come to develop

reputations that precede them and warn potential victims to avoid provoking the covert

aggressor.









APPENDIX A
PERSONALITY PACKET (PART 1)

Carefully read each section's instructions.

Questions 1-33: Listed below are statements concerning your attitudes, traits, and tendencies.
Read each item and decide whether the statement is true or false as it pertains to you.
1 = True
2 = False
1. Before voting I thoroughly investigate the qualifications of all the candidates.
2. I never hesitate to go out of my way to help someone in trouble.
3. It is sometimes hard for me to go on with my work if I am not encouraged.
4. I have never intensely disliked anyone.
5. On occasion I have had doubts about my ability to succeed in life.
6. I sometimes feel resentful when I don't get my way.
7. I am always careful about my manner of dress.
8. My table manners at home are as good as when I eat out in a restaurant.
9. If I could get into a movie without paying and be sure I was not seen, I would probably do it.
10. On a few occasions, I have given up doing something because I thought too little of my ability.
11. I like to gossip at times.
12. There have been times when I felt like rebelling against people in authority even though I
knew they were right.
13. No matter who I'm talking to, I'm always a good listener.
14. I can remember "playing sick" to get out of something.
15. There have been occasions when I took advantage of someone.
16. I'm always willing to admit it when I make a mistake.
17. I always try to practice what I preach.
18. I don't find it particularly difficult to get along with loud-mouthed, obnoxious people.
19. I sometimes try to get even, rather than forgive and forget.
20. When I don't know something I don't at all mind admitting it.
21. I am always courteous, even to people who are disagreeable.
22. At times I have really insisted on having things my own way.
23. There have been occasions when I felt like smashing things.
24. I would never think of letting someone else be punished for my wrongdoings.
25. I never resent being asked to return a favor.
26. I have never been irked when people expressed ideas very different from my own.
27. I never make a long trip without checking the safety of my car.
28. There have been times when I was quite j ealous of the good fortune of others.
29. I have almost never felt the urge to tell someone off.
30. I am sometimes irritated by people who ask favors of me.
31. I have never felt that I was punished without cause.
32. I sometimes think when people have a misfortune they only got what they deserved.
33. I have never deliberately said something that hurt someone's feelings.










Questions 34-62: Using the 5 point scale shown below, indicate how uncharacteristic or
characteristic each of the following statements is in describing you.

1 = extremely uncharacteristic of me
2 = somewhat uncharacteristic of me
3 = neither uncharacteristic or characteristic of me
4 = somewhat characteristic of me
5 = extremely characteristic of me

34. Some of my friends think I am hotheaded
35. If I have to resort to violence to protect my rights, I will.
36. When people are especially nice to me, I wonder what they want.
37. I tell my friends openly when I disagree with them.
38. I have become so mad I have broken thinks
39. I can't help getting into arguments when people disagree with me.
40. I wonder why sometimes I feel so bitter about things.
41. Once in a while, I can't control the urge to strike another person.
42. I am an even-tempered person.
43. I am suspicious of overly friendly strangers.
44. I have threatened people I know.
45. I flare up quickly but get over it quickly
46. Given enough provocation, I may hit another person.
47. When people annoy me, I may tell them what I think of them.
48. I am sometimes eaten with j ealousy.
49. I can think of no good reason to ever hitting a person.
50. At times, I feel I have gotten a raw deal out of life.
51. I have trouble controlling my temper.
52. When frustrated, I let my irritation show.
53. I sometimes feel that people are laughing at me behind my back.
54. I often find myself disagreeing with people.
55. If somebody hits me, I hit back.
56. I sometimes feel like a powder keg ready to explode
57. Other people always seem to get the breaks.
58. There are people who pushed me so far that we came to blows.
59. I know that "friends" talk about me behind my back.
60. My friends say that I'm somewhat argumentative.
61. Sometimes I fly off the handle for no good reason.
62. I get into fights a little more than the average person.










Questions 63-88: Here are a number of characteristics that may or may not apply to you. For
example, do you agree that you are someone who likes to spend time with others? Please write a
number next to each statement to indicate the extent to which you agree or disagree with that
statement.

Disagree Disagree Neither agree Agree Agree
strongly a little nor disagree a little strongly
1 2 3 4 5

I See Myself as Someone Who...

63. Is talkative
64. Does a thorough j ob
65. Is original, comes up with new ideas
66. Is reserved
67. Can be somewhat careless
68. Is curious about many different things
69. Is full of energy
70. Is a reliable worker
71. Is ingenious, a deep thinker
72. Generates a lot of enthusiasm
73. Tends to be disorganized
74. Has an active imagination
75. Tends to be quiet
76. Tends to be lazy
77. Is inventive
78. Perseveres until the task is finished
79. Values artistic, aesthetic experiences
80. Is sometimes shy, inhibited
81. Does things efficiently
82. Prefers work that is routine
83. Is outgoing sociable
84. Makes plans and follows through with them
85. Likes to reflect, play with ideas
86. Has few artistic interests
87. Is easily distracted
88. Is sophisticated in art, music, or literature










Questions 89-116: How frequently in the last 6 months might you engage in the behaviors
listed below when you are angry with someone? Use the scale below.

1 2 3 4 5
Never Very Often

89. Yelled or screamed at them
90. Did things to irritate them
91. threatened to hit or throw something at them
92. made up stories to get them in trouble
93. did not show that I was angry
94. cursed at them
95. threw something at them
96. tried to make them look stupid
97. stomped out of the room
98. made negative comments about their appearance to someone else
99. hit (or tried to hit) them with something hard
100. insulted them or called them names to their face
101. talked the matter over
102. spread rumors about them
103. sulked and refused to talk about it
104. kicked (or tried to kick) the other person
105. dropped the matter entirely
106. took something that belonged to them
107. hit (or tried to hit) the other person but not with anything
108. gossiped about them behind their back
109. pushed, grabbed, or shoved them
110. called them names behind their back
111. told others not to associate with them
112. waited until I calmed down and then discussed the problem
113. told others about the matter
114. threw something (but not at the other) or smashed something
115. destroyed or damaged something that belonged to them
116. gathered other friends to my side










APPENDIX B
PERSONALITY PACKET (PART 2)

The following statements inquire about your thoughts and feelings in a variety of situations. For
each item, indicate how well it describes you by choosing the appropriate number on the scale at
the top of the page. Answer as honestly as you can. Thank you.
1 2 3 4 5
Does not Describes
describe me very
me well well

1. I often put off until tomorrow things I know I should do today.
2. Before criticizing somebody, I try to imagine how I would feel if I were in their place.
3. I am basically a lazy person.
4. I find satisfaction in working as well as I can.
5. I prefer to work in situations that require a high level of skill.
6. I more often attempt tasks that I am not sure I can do than tasks I know I can do.
7. If I'm sure I'm right about something, I don't waste much time listening to other people's arguments.
8. If I'm not good at something, I would rather keep struggling to master it than move on to something I
may be good at.
9. I find satisfaction in exceeding my previous performance even if I don't outperform others.
10. I like to work hard.
11. I sometimes try to understand my friends better by imagining how things look for their perspective.
12. I believe that there are two sides to every question and try to look at them both.
13. I sometimes find it difficult to see things from the "other guy's" point of view.
14. Hard work is something I like to avoid.
15. I try to look at everybody's side of a disagreement before I make a decision.
16. I hate to see bad workmanship.
17. When I'm upset at someone, I usually try to "put myself in his shoes" for a while.
18. I can easily sit for a long time doing nothing.
19. I must admit I often do as little work as I can get away with.
20. I easily get bored if I don't have something to do.
21. When I see someone being taken advantage of, I feel kind of protective toward them.
22. I get a sense of satisfaction out of being able to say I have done a very good job on a project.
23. I would rather lean easy fun games than difficult thought games.
24. When I see someone being treated unfairly, I sometimes don't feel very much pity for them.
25. I often have tender, concerned feelings for people less fortunate than me.
26. It is no use playing a game when you are playing with someone as good as yourself.
27. I would describe myself as a pretty soft-hearted person.
28. Sometimes I don't feel very sorry for other people when they are having problems.
29. Other people's misfortunes do not usually disturb me a great deal.
30. There is satisfaction in a job well done.
31. I am often quite touched by things that I see happen.
32. Part of the satisfaction in doing something comes from seeing how good the finished product looks.
33. I would rather do something at which I feel confident and relaxed than something which is
challenging and difficult.
34. I like to be busy all the time.
35. I feel like giving up quickly when things go wrong.









APPENDIX C
BASIC INFORMATION QUESTIONNAIRE



First Name: Taylor

Gender: Male Female Age: 21

Why are you at the University of Florida and what are your post-graduation plans?
I am a senior here at UF. I came to the University of Florida because of its good academic
reputation at a pretty good price. I hope to attend graduate school next year after I graduate in
May. I'm taking intro psych right now with Dr. Miller and I really like it so far. I'm
thinking about taking another psychology class next semester.



What do you like to do for fun?
I enj oy hanging out with my friends during the weekend.










APPENDIX D
FEEDBACK

Evaluator First Name: Taylor

Form: S qui rrel-Proof B irdfeeder
GPACPL Form B-2.7

Time-limit: 6 minutes

Please take a moment to review the problem assigned to your fellow participant. Make sure that
you fully understand what was described in the instructions. If you have any questions, please
ask the researcher now.

Once you understand the problem assigned to your fellow participant, take a few moments to
assess the solution developed by him/her in response to this problem.

In the space below, please give your impression of this participant's solution. Refer to the
instructions associated with the participant's specific form of the test to determine how well
his/her solution solves the problem.

Positive Feedback:
This design is a pretty creative one. I think it has a lot of potential to solve the squirrel proof
birdfeeder in a useful and inventive way. Overall, good job. This person probably has a lot of
creative intelligence.

Negative Feedback:
This design is pretty bad. I don't think it would really work well, if at all. I can easily think of
several different solutions that would work much better. This person probably doesn't have much
creative intelligence.

Ambiguous Feedback:
This person's abilities are self evident. A thorough analysis of his/her performance will surprise
you. This person's worth to his/her future employer can only be imagined. This person has
relatively few minor weaknesses. There is no doubt as to the extent of this person' s creative
intelligence .









APPENDIX E
FIRST IMPRESSIONS QUESTIONNAIRE (MIXED MESSAGES CONDITION)

***Condition: Mixed Messages. xi denotes a mixed messages statement, N denotes a negative
message statement and P denotes a positive message statement.
In real world groups, people work with individuals about whom they have varying degrees of knowledge,
ranging from barely knowing some to knowing others well. In all cases, the impressions people have of
others can be important. Some impressions may be held with confidence whereas others may be little
more than guesses or hunches. Depending on the condition to which you've been assigned, you may have
exchanged more or less information with the other, and this may also affect your impressions and the
confidence you have in them. Moreover, people are often aware of what others think of them before ever
meeting them, so we need to take this into account.

This questionnaire assesses your impressions. In the condition to which you have been assigned, your
impressions will be shown to the other person prior to the second assessment. However, you will not learn
the other person's impressions of you until the end of the session.

Please give us your impression of the other person, even if you are only guessing. Therefore, please
review the statements below that describe individuals in a variety of domains-interpersonal, task
performance, etc. From the list, please select 8 statements that you believe best describe the other person
by placing an "X" on the 1st blank line next to the statement. Then, rank each of the ten selected
statements from 1 to 8 where 1 most descriptive, 2 2nd mOst descriptive, 3 3rd mOst descriptive, etc.,
through 8= 8th mOst descriptive by writing a number 1-8 on the second blank line next to the statement.
The statements you did not select should all be less descriptive of the person than the eight that you
selected.

Select Rank
by writing selected
an X statements

The quality of this person' s work is well-known. NT
Several people actively dislike this person.
This person is someone I would like to get to know better. P
This is not the kind of person you encounter everyday. NT
This person is not easy to respect.
This person is capable of doing many things. P
This person is quite distinct from his/her peers. NT
This person possesses many positive qualities. P
This person is hard to get to know.
This person probably has a well-known reputation. xi
A lot of people would like to be this person' s close friend. P
This person finds novel tasks difficult. N
This person is a well-rounded individual. P
This person has made quite an impression on me. xi
Few people consider this person one of their close friends.
Many people come to this person to seek personal advice. P
This person is quite interesting.
People usually find this person to have an unpleasant personality.









APPENDIX F
FIRST IMPRESSIONS QUESTIONNAIRE (NEUTRAL MESSAGES CONDITION)

***Condition: Neutral Messages NL denotes a neutral message statement, N denotes a negative
message statement and P denotes a positive message statement.***

In real world groups, people work with individuals about whom they have varying degrees of knowledge,
ranging from barely knowing some to knowing others well. In all cases, the impressions people have of
others can be important. Some impressions may be held with confidence whereas others may be little
more than guesses or hunches. Depending on the condition to which you've been assigned, you may have
exchanged more or less information with the other, and this may also affect your impressions and the
confidence you have in them. Moreover, people are often aware of what others think of them before ever
meeting them, so we need to take this into account.

This questionnaire assesses your impressions. In the condition to which you have been assigned, your
impressions will be shown to the other person prior to the second assessment. However, you will not learn
the other person's impressions of you until the end of the session.

Please give us your impression of the other person, even if you are only guessing. Therefore, please
review the statements below that describe individuals in a variety of domains-interpersonal, task
performance, etc. From the list, please select 8 statements that you believe best describe the other person
by placing an "X" on the 1st blank line next to the statement. Then, rank each of the ten selected
statements from 1 to 8 where 1 most descriptive, 2 2nd mOst descriptive, 3 3rd mOst descriptive, etc.,
through 8= 8th mOst descriptive by writing a number 1-8 on the second blank line next to the statement.
The statements you did not select should all be less descriptive of the person than the eight that you
selected.

Select Rank
by writing selected
an X statements

This person has taken a formal logic course. NL
Several people actively dislike this person.
This person is someone I would like to get to know better. P
This person probably has pets. NL
This person is not easy to respect.
This person is capable of doing many things. P
He/she lives off campus. NL
This person possesses many positive qualities. P
This person is hard to get to know.
This person comes from a large family. NL
A lot of people would like to be this person' s close friend. P
This person finds novel tasks difficult. N
This person is a well-rounded individual. P
He/she wears corrective lenses. NL
Few people consider this person one of their close friends.
Many people come to this person to seek personal advice. P
This person is originally from Florida. NL
People usually find this person to have an unpleasant personality.










APPENDIX G
PRELIMINARY INTERACTION ASSESSMENT

Instructions. This scale consists of a number of words that describe different feelings and
emotions. Read each item and then mark the appropriate answer in the space next to that word.
Indicate to what extent you feel this way right now, at this moment. Use the following scale to
record your answers:


1............ ..... ...... .2 .... ........ .. . . . . . . .
very slightly a little moderately quite a bit extremely
or not at all

It is very important that you answer as accurately as possible to ensure the validity of the final
assessment. Your responses to these questions will not be shown to the other participant.


interested irritable

distressed alert

excited ashamed

upset inspired

strong nervous

guilty determined

scared attentive

hostile jittery


enthusiastic

proud


active

afraid










Please answer the following questions by circling the number that best corresponds to your opinion.
Depending on the condition to which you have been assigned, some of the questions may seem
irrelevant because you don't expect to meet your partner. If you are assigned to such a condition,
imagine how a meeting with your partner would be.
It is very important that you answer as accurately as possible to ensure the validity of the final
assessment. Your responses to these questions will not be shown to the other participant.

1. Do you believe that you will like the other participant?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit

2. Do you think that the other participant is generally friendly?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit

3. Do you believe you will respect the other participant?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit

4. How much do you think you will enjoy working with the other participant?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit

5. How well do you think this person gets along with other people?

1........2... .....3...... 4........5... .....6...._...7........8........9
not at all not very somewhat fairly extremely
well well well well well

6. Do you believe that the other participant will like you?

1........2........3...... 4........5........6........7... .....8........9
not not somewhat a good extremely
at all much bit

7. How much do you think the other participant respects you?

1........2........3...... 4........5........6........7... .....8........9
not not somewhat a good extremely
at all much bit










8. How much creative intelligence do you think the other participant possesses?


1........2........3......
not at all not very
creatively creatively
intelligent intelligent


4........5........6........7........8... .9
somewhat fairly extremely
creatively creatively creatively
intelligent intelligent intelligent


9. How much do you value the other participant' s opinion regarding creative intelligence
problem solving?
1........2........3...... 4........5........6........7... .....8........9
not not somewhat a good extremely
at all much bit

10. How much creative intelligence do you believe you possess?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit

11. How much creative intelligence does the other participant believe you have?


1........2........3......
not at all not very
creatively creatively
intelligent intelligent


4........5........6........7........8... .9
somewhat fairly extremely
creatively creatively creatively
intelligent intelligent intelligent


12. How much creative intelligence do you think the other participant believes he/she has?


1........2........3......
not at all not very
creatively creatively
intelligent intelligent


4........5........6........7........8... .9
somewhat fairly extremely
creatively creatively creatively
intelligent intelligent intelligent


13. How well do you think the first assessment of creative intelligence actually assessed
creative intelligence?

1........2........3...... 4........5........6........7... .....8........9
not at all not very somewhat fairly extremely
well well well well well

14. How well do you think the second assessment of creative intelligence will assess your
creative intelligence?

1........2........3....... 4........5........6........7........8... .9
not at all not very somewhat fairly extremely
well well well well well










15. How clearly do you express your thoughts and feelings to others?

1........2........3....... 4........5........6........7........8... .9
not at all not very somewhat fairly extremely
Well well well well well

16. How well do you understand the thoughts and feelings that others express to you?

1........2........3...... 4........5........6........7... .....8........9
not at all not very somewhat fairly extremely
Well well well well well

17. To what extent do you prefer working with others versus working alone?

1........2........3....... 4........5........6........7........8... .9
always or mostly with others mostly always or
Almost always alone and alone with almost always
alone equally others with others

18. To what extent do you think your final answer on the second assessment will be equal
parts your work and the other participant' s work?

1........2........3......4........5........6........7... .....8........9
entirely mostly equally mostly entirely
other other me and me me
person person other
person

19. To what extent do you think the second assessment will be difficult?

1........2........3...... 4........5........6........7... .....8........9
not not somewhat a good extremely
at all much bit

20. How positively did you evaluate the other person's work?

1........2........3...... 4........5........6........7... .....8........9
not at all not very somewhat fairly Extremely
at all very positive Positively Positively
Positively positively

21. How positively did the other person evaluate your work?

1........2........3....... 4........5........6........7........8... .9
not at all not very somewhat fairly Extremely
at all very positive Positively Positively
Positively positively











22. To what extent do you think that the other person is straightforward?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit


23. To what extent do you think that the other person is sincere?

1........2........3...... 4........5........6........7... .....8........9
not not somewhat a good extremely
at all much bit

24. To what extent do you think that the other person is a skilled communicator?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit


25. How much are you worried about interacting with the other person?

1........2........3....... 4........5........6........7........8... .9
not not somewhat a good extremely
at all much bit













Table H-1. Statement selection means
Feedback Statement type Mixed messages Neutral messages
Ambiguous feedback
Alternative 3.09ab (0.70) 2.57, (0.83)
messages
Negative messages 0.07de (0.33) 0.03df (0.16)
Positive messages 4.84, (0.77) 5.41h, (0.86)
Negative feedback
Alternative 3.34aij (1.01) 3.31i (0.90)
messages
Negative messages 1.83 (1.51) 2.83 (1.76)
Positive messages 2.83, (1.93) 1.86 (1.52)
Positive feedback
Alternative 2.79bk (0.80) 2.79ck (0.82)
messages
Negative messages 0.05e1 (0.22) 0.05fl (0.05)
Positive messages 5.15,m (0.87) 5.16hm (0.90)
Values reflect average number of messages selected out of a possible 6. Orthogonal simple
effects comparisons are presented in the table, with means having at least one factor in common
and sharing a subscript not significantly different from one another. Standard deviations are
indicated in parentheses.


APPENDIX H
STATEMENT SELECTION MEANS









APPENDIX I
STATEMENT SELECTION MEANS WITH OFFSET COMPARISON GROUPS NOT
EXPECTING A FUTURE INTERACTION

Table I-1. Statement selection means with offset comparison groups not expecting a future
interaction
Feedback Statement type Future meeting No future meeting
Ambiguous feedback
Mixed messages 3.09ab (0.70) 2.97a (0.81)
Negative messages 0.07, (0.33) 01, (0.32)
Positive messages 4.84d (0.77) 4.92d (0.73)
Negative feedback
Mixed messages 3.34bef (1.01) 3.74e (1.12)
Negative messages 1.83, (1.51) 2.10, (1.27)
Positive messages 2.83fh, (1.93) 2.16h, (1.37)
Values reflect average number of messages selected out of a possible 6. Orthogonal simple
effects comparisons are presented in the table, with means having at least one factor in common
and sharing a subscript not significantly different from one another. Standard deviations are
indicated in parentheses.










LIST OF REFERENCES


Archer, J., & Coyne, S. M. (2005). An integrated review of indirect, relational, and social
aggression. Personality and' Social Psychology Review, 9 (3), 212-230.

Archer, J., & Webb, I. A. (2006). The relation between scores on the Buss-Perry Aggression
Questionnaire and aggressive acts, impulsiveness, competitiveness, dominance, and
sexual jealousy. Aggressive Behavior, 32, 464-473.

Baron, R., & Richardson, D. R. (1994). Human Aggression. New York: Plenum.

Berglas, S., & Jones, E. E. (1978). Drug choice as a self-handicapping strategy in response to
noncontingent success. Journal ofPersonality and' Social Psychology, 36 (4), 405-417.

Bjiirkqvist, K., Lagerspetz, K., Kaukiainen, A. (1992). Do girls manipulate and boys fight?
Developmental trends in regard to direct and indirect aggression. Aggressive Behavior,
18, 117-127.

Bjiirkqvist, K., Ojsterman, K., & Lagerspetz, K. M. J. (1994) Sex differences in covert aggression
among adults. Aggressive Behavior, 20, 27-33.

Bjiirkqvist, K., Ojsterman, K., Lagerspetz, K., Landau, S. F., Caprara, G., & Fraczek, A. (2001).
Aggression, victimization and sociometric status: Findings from Finland, Israel, Italy and
Poland. In J. M. Rameriez & D. S. Richardson (Eds.), Crosscultural approaches to
aggress.v~ion and' reconciliation (pp. 111-119). Huntington, NY: Nova Science.

Bradbury, T.N., & Fincham, F.D. (1992). Attributions and behavior in marital interaction.
Journal of personality and' Social Psychology, 63, 613-628.

Bureau of Justice Statistics Crime Characteristics (2007). Retrieved April 2, 2007, from
http://www.oj p.usdoj .gov/bj s/cvict~c.htm#relate

Bushman, B. J., & Anderson, C. A. (2001). Is it time to pull the plug on hostile versus
instrumental aggression dichotomy? Psychological Review, 108 (1), 273-279.

Buss, A. H. (1961). The Psychology ofAggression. New York: John Wiley and Sons, Inc.

Buss, A. H., & Perry, M. (1992). The Aggression Questionnaire. Journal ofPersonality and'
Social Psychology, 63(3), 452-459.

Cairns, R. B., Cairns, B., Neckerman, H., Ferguson, L., & Gariepy, J. (1989). Growth and
aggression: 1. Childhood to early adolescence. Developmental Psychology, 25, 320-330.

Campos, B., Keltner, D., Beck, J.M., Gonzaga, G. C., & John, O.P. (2007). Culture and teasing:
The relational benefits of reduced desire for positive self-differentiation. Personality and'
Social Psychology Bulletin, 33 (1), 3-16.










Cohen, D., Vandello, J., Puente, S., & Rantilla, A. (1999). "When you call me that, smile!" How
norms for politeness, interaction styles, and aggression work together in southern culture.
Social Psychology Quarterly, 62 (3), 257-275.

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression /
correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence
Erlbaum.

Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social-information
processing mechanisms in children's social adjustment. Psychological Bulletin, 115 (1),
74-101.

Cri ck, N. R., & Grotpeter, J. ( 199 5). Rel ati onal aggressi on, gender, and soci al-p sychol ogi cal
adjustment. Child Development, 66, 710-722.

Crick, N. R., Werner, N., Casas, J., O'Brien, K., Nelson, D., Grotpeter, J., et al. (1999).
Childhood aggression and gender: A new look at an old problem. Nebraska Symposium
on Motivation, 45, 75-141.

Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of
psychopathology. Journal of Consulting Psychology, 24, 349-354.

Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS
Catalog of Selected Documents in Psychology, 10, 85.

Gaertner, S. L. & Dovidio, J. F. (1977). The subtlety of white racism, arousal, and helping
behavior. Journal ofPersonality and' Social Psychology, 35 (10), 691-707.

Galen, B. R., & Underwood, M. K. (1997). A developmental investigation of social aggression
among children. Developmental Psychology, 33, 589-600.

Geen, R. G., Stonner, D., & Shope, G. L. (1975). The facilitation of aggression by aggression:
Evidence against the catharsis hypothesis. Journal ofPersonality and' Social Psychology,
31(4),721-726.

Green, L. R., Richardson, D. R., & Lago, T. (1996). How do friendship, indirect, and direct
aggression relate? Aggressive Behavior, 22, 81-86.

Goldstein, S. E., Tisak, M. S., Persson, A. V., & Boxer, P. (2006). Children's evaluations of
ambiguous provocation by relationally aggressive, physically aggressive, and prosocial
peers. British Journal ofDevelopmental Psychology, 24 (4), 701-708.

Gordon, R. A. (1996) Impact of ingratiation on judgments and evaluations: A meta-analytic
investigation. Journal ofPersonality and Social Psychology, 71 (1), 54-70.

Hardin, C. D., & Higgins, E. T. (1996). Shared reality: How social verification makes the
subjective objective. In R. M. Sorrentino & E. T. Higgins (Eds.), Hand'book of motivation
and' cognition: Vol 3. The interpersonal context, (pp. 28-84). New York: Guilford Press.










Jones, E. E. (1964). Ingratiation. New York: Appleton-Century-Crofts .


Kaukiainen, A., Bjoirkqvist, K., Lagerspetz, K., Ojsterman, K., Salmivalli, C., Rothberg, S., &
Ahlbom, A. (1999). The relationships between social intelligence, empathy, and three
types of aggression. Aggressive Behavior, 25, 81-89.

Klar, Y., & Giladi, E. E. (1997). No one in my group can be below the group's average: A robust
positivity bias in favor of anonymous peers. Journal of Personality and Social
Psychology, 73(5), 885-901.

Kowalski, R. M., Howerton, E., & McKenzie, M. (2001). Permitted disrespect: Teasing in
interpersonal interactions. In R. M. Kowalksi (Ed.), Behaving Badly: Aversive Behaviors
in Interpersonal Relationships (pp. 177-202), Washington, DC: American Psychological
Association.

LaFontana, K. M., & Cillessen, A. H. N. (2002). Children's perception of popular and unpopular
peers: A multimethod assessment. Developmental Psychology, 38 (5), 63 5-647.

Lagerspetz, K., Bjoirkqvist, K., & Peltonen, T. (1988). Is indirect aggression typical of females?
Gender differences in aggressiveness in 11- to 12-year old children. Aggressive Behavior.
14, 403-414.

Miller, M. L. & Schlenker, B. R. (2008). [Disguising aggression between friends]. Unpublished
raw data.

Million, T., Davis, R., & Millon, C. (1997) M~CMI-III: M~illon Clinical2\~ultia~xiallnventory-III
Manual. 2nd ed. Minneapolis, MN: NCS Pearson, Inc.

Richardson, D. R., & Green, L. R. (2003). Defining direct and indirect aggression: The
Richardson Conflict Response Questionnaire. Revue Internationale de Psychologie
Sociale, 16(2), 1 1-3 0.

Salmivalli, C., Kaukiainen, A., & Lagerspetz, K. (2000). Aggression and sociometric status
among peers: Do gender and type of aggression matter? Scandinavian Journal of
Psychology, 41, 17-24.

Schlenker, B. R. (1980). Impression management: The self-concept, social identity, and
interpersonal relations. Monterey, CA: Brooks/Cole.

Schlenker, B. R. (2003). Self-presentation. In M. R. Leary and J. P. Tangney (Eds.), H~andbook
ofselfand identity (pp. 492-518). New York: Guilford.

Schlenker, B. R., & Britt, T. W. (1999). Beneficial impression management: Strategically
controlling information to help friends. Journal ofPersonality and Social Psychology,
76, 559-573.










Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability.
Psychological Bulletin, 86 (2), 420-428.

Snyder, M. L., Kleck, R. E., Strenta, A., & Mentzer, S. J. (1979). Avoidance of the Handicapped:
An attributional ambiguity analysis. Journal ofPersonality and' Social Psychology, 3 7
(12), 2297-2306.

Tedeschi, J. T., & Felson, R. B. (1994). Violence, Aggression, and' Coercive Actions.
Washington, DC: American Psychological Association.

Vonk, R. (2002). Self-serving interpretations of flattery: Why ingratiation works. Journal of
Personality and' Social Psychology, 82 (4), 515-526.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: The PANAS Scales. Journal of Personality and' Social
Psychology, 54, 1063-1070.









BIOGRAPHICAL SKETCH

Marisa L. Miller was born in Valdosta, Georgia in 1981. She grew up in various cities in

Georgia, graduating from Troup County Comprehensive High School in LaGrange, Georgia in

1999. She graduated summa cum laude with honors from Mercer University in Macon, Georgia

in 2003 with a Bachelor of Arts in psychology and dual minors in art and photography. In fall

2003, she began her graduate work at the University of Florida were she earned her Master of

Science in psychology, before beginning her doctoral work in social psychology.





PAGE 1

1 COVERT AGGRESSION: THE MEANS AND MOTIVE OF GETTING AWAY WITH IT By MARISA L. MILLER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Marisa L. Miller

PAGE 3

3 ACKNOWLEDGMENTS I am immensely thankful to the many people who supported me throughout this process. First, I would like to thank my supervisory committee chair, Dr Barry Schlenker. From the theoretical seeds of this project, he helped nourish these ideas and my excitement for them. With endless patience and vast wisdom, he helped me make this project what it is. In this project and throughout my graduate training, he broadened my knowledge and deepened my understanding of scientific psychological resear ch, and for that I am immensely grateful. I would also like to thank my other supervisory committee members. I th ank Dr. Ken Rice for hi s point of view that helped broaden my thinking. I thank Dr. James Algina for his endless willingness to answer statistical questions, and to help me navigate the often frustrating world of SAS. I would also like to especially thank Dr. James Shepperd for his se nse of humor, his encouragement of these ideas in their initial stages, and his honest enthusiasm for this project. I would also like to thank my friends and fam ily who gave me the s upport I needed. To my parents, Kathy Davis and Bill Miller, I thank them for instilling in me an interest in thinking about the world, and a fascination with understandi ng and thinking critically about the behavior of people in this world. To my graduate sc hool friends, Jodi Grace, Penny McNatt Devine, Melinda Goodman, Meredith Terry, Kate Swee ny, Wendi Malone, Gizem Aksoy Toska, Ryan Johnson, and others, I thank them more th an they know for their immense support, commiseration, humor, empathy, and friends hipalways showing perfect understanding, sometimes without knowing that it was needed. I would also like to thank my longtime friends and extended family, Kristen Miller, Will Miller, Barry Davis, Katy Nichols, Erin Zier, Venussa Kassees, Maryliz Pinto, Nessa Manten and others, who gave me a fresh dose of reality and kept me grounded in the real world out side of graduate school, and showed unrestrained and constant

PAGE 4

4 faith in my ability to succee d. I particularly thank Katy Nichols for enduring with me the experiences that inspired this project. Finally, I would like to thank coffee. Despite my initial distaste for coffee 5 years ago, I grew to crave its caffeinated goodness. I would not have survived graduate school or this dissertation process without it.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT.....................................................................................................................................9 CHAP TER 1 INTRODUCTION..................................................................................................................11 Aggression Taxonomy............................................................................................................ 12 Physical versus Verbal....................................................................................................12 Active versus Passive......................................................................................................13 Direct versus Indirect...................................................................................................... 16 Special Note on Terminology: Indirect, Relational, and Social Aggression ................... 17 Gender Differences.................................................................................................................19 Intelligence and Indirect Aggression...................................................................................... 20 Self-Presentation and Reputati on in Aggressions Afterm ath................................................24 Situational Factors............................................................................................................ ......31 Hypotheses..............................................................................................................................33 Current Study..........................................................................................................................40 2 METHOD......................................................................................................................... ......42 Overview....................................................................................................................... ..........42 Design.....................................................................................................................................42 Participants.............................................................................................................................43 Prescreening Session..............................................................................................................43 3 RESULTS...............................................................................................................................49 Pilot Testing............................................................................................................................49 Feedback..........................................................................................................................49 Message Options.............................................................................................................50 Scoring the First Impr essions Questionnaire .......................................................................... 51 Overview of Analyses.............................................................................................................53 Messages Used........................................................................................................................54 Positive, Negative and Alternate Messages..................................................................... 54 Category Comparison......................................................................................................56 Summary of Message Use...............................................................................................60 Objective Assessment of Design and Evaluation Quality............................................... 63 Interaction Expectations....................................................................................................... ..64

PAGE 6

6 Perceptions of Partner......................................................................................................... ....66 Affect......................................................................................................................................69 Communication Preference....................................................................................................70 Moderators..................................................................................................................... .........70 Perspective Taking.......................................................................................................... 71 Other Moderators.............................................................................................................72 4 DISCUSSION.........................................................................................................................87 Comparisons between Positiv e and Am biguous Feedback.................................................... 89 Nature of the Interp ersonal Relationship ................................................................................94 Gender.....................................................................................................................................96 Expectation of a Future Interaction........................................................................................ 98 Social Desirability............................................................................................................ ....101 Perspective Taking............................................................................................................. ...102 Distinguishing Covert Aggression from Other Forms of Aggression.................................. 103 APPENDIX A PERSONALITY PACKET (PART 1)..................................................................................106 B PERSONALITY PACKET (PART 2)..................................................................................110 C BASIC INFORMATION QUESTIONNAIRE.................................................................... 111 D FEEDBACK.........................................................................................................................112 E FIRST IMPRESSIONS QUESTIONNAIRE (MIXED MESSAGES CONDITION) ......... 113 F FIRST IMPRESSIONS QUESTIONNAIRE (NE UTRAL MESSAGES CONDITION).... 114 G PRELIMINARY INTERACTION ASSESSMENT............................................................. 115 H STATEMENT SELECTION MEANS.................................................................................120 I STATEMENT SELECTION MEANS WI T H OFFSET COMPARISON GROUPS NOT EXPECTING A FUTURE INTERACTION............................................................... 121 LIST OF REFERENCES.............................................................................................................122 BIOGRAPHICAL SKETCH.......................................................................................................126

PAGE 7

7 LIST OF TABLES Table page 3-1. Pilot study: Perceptions of feedback.................................................................................. 75 3-2. Ranking means...................................................................................................................76 3-3. Category ranking difference scores................................................................................... 77 3-4. Ranking means with offset comparison gr oups not expecting a future interaction ........... 78 3-5. Trait groupings and relevant Cronbach alphas.................................................................. 79 3-6. Perceptions of partner and affect....................................................................................... 80 3-7. Participant communication m ethod preferences................................................................81 3-8. Ranking of negative messages as a func tion of perspective taking, feedback, and response option ..................................................................................................................82 3-9. Ranking of positive and negative messages as a f unction of social desirability, feedback, and response option...........................................................................................83 H-1. Statement selection means............................................................................................... 120 I-1. Statement selection means with offset com parison groups not expecting a future interaction........................................................................................................................121

PAGE 8

8 LIST OF FIGURES Figure page 3-1. Mean ranking for negative m essages as a function of perspective taking, feedback, and response option............................................................................................................84 3-2. Mean ranking for negative m essages as a f unction of social desirability, feedback, and response option............................................................................................................85 3-3. Mean rank ing for positive messages as a function of social desirability, feedback, and response option............................................................................................................86

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy COVERT AGGRESSION: THE MEANS AND MOTIVE OF GETTING AWAY WITH IT By Marisa L. Miller August 2008 Chair: Barry R. Schlenker Major: Psychology Existing aggression literature covers a variety of aggressive means and motives, but so far little research has examined the nature of c overt aggressionaggressi on that simultaneously contains both a positive and negative interp retation, thereby possibly escaping the negative consequences of such aggression. This study stands as the first experimental investigation of responses to covert aggressi on and the use of covert aggression. Participants (N = 295) completed a bogus creative intell igence assessment and received false feedback from a partner (actually nonexistent). The evaluative feedback was either positive, negative, or conveyed an ambiguous (mixed) message. Partic ipants then had the opportunity to send a combination of positive, negative, and alternate messages to the evaluator. The alternate messages were either ambiguous, and thereby provided an opportunity to employ covert aggression, or neutral, and thereby provided no opportunity fo r covert aggression. Participants subsequently reported their perceptions of their partners and emotional state. Pilot participants verified that the feedback in the ambiguous, positive and negative feedback we re perceived as containing mixed, positive, and negative interpretations, respectively. In general, participants did not respond negatively to the ambiguous feedback and viewed ambiguous and positive feedback partners in a similar, positive manner. In contrast, participants receiving negative feedback were subtly affected by the

PAGE 10

10 availability of mixed messages, indicating that when people can send mixed messages, they do so and are, on the surface at least, nicer than when they must express hostility in an open and clear manner. Negative feedback participants expe rienced more negative affect and reported less favorable views of their partne rs than either positive or am biguous feedback participants. Individual difference measures mode rating these effects are discussed.

PAGE 11

11 CHAPTER 1 INTRODUCTION If you cant say anyth ing nice, then at least have the decency to be vague. Unknown History books chronicle the existence of hu mankinds aggression pi tting one civilization against another with history-changing consequences. Without a doubt, wars, shots fired, and men killed have certainly had long-la sting impact on societies throughout the world. However, just as many epic outcomes have resulted from far less conspicuous conflicts. Governments, militaries, and political groups have used spies throughout history to garner information, complete assassinations, and to aggress against other governments, militaries, and political groups in a covert manner that allows the sponsoring agent to preserve an image of dutifully abiding by formal agreements. Government spies frequently a dopt and portray alternate identities that gain the confidence of others, gain acc ess to important information, a nd conceal the spies ulterior motives. Although perhaps less likely to be recorded in history books, this subversive and covert form of approaching hostility and aggression that maintains a surf ace appearance devoid of malicious actions is just as likely to have impor tant consequences for political groups as more conspicuous and physically violent actions. In a similar manner, in dividuals in everyday life also desire the appearance of civility while secr etly devising and enacting hostile activities. Maintaining the appearance of one thing while doing another is a useful but potentially dangerous path to choose in interpersonal rela tionships. Unfortunately, lit tle to no research has directly assessed interpersona l aggression considering this covert motive. Instead, much aggression research has ignored identity concernshow the would-be aggressor appears to

PAGE 12

12 others and manages his/her identity in the aftermath of aggression. In short, he/she must actively disguise his/her aggressive actions by engaging in covert aggression. Although aggression comes in many forms a nd guises, social ps ychologists define aggression as behavior th at is intended to cause harm. Pr evious research highlights several dimensions by which aggression can be categor ized and also pinpoint s specific personal characteristics and situational determinants of aggression. While some research delves into territory and behaviors that may be dually charac teristic of the covert and overt aggressor, a covert aggressor faces challenges distinct from those of the overt aggressor, and there is reason to conclude that unique processe s are at work as the covert aggr essor acts in his/her social world. To fully understand the unique struggles and cons iderations of the cove rt aggressor, it is necessary to first have a proper understanding of the more prevailing aggression research before focusing on indirect aggression, a similar but still distinctly different type of aggression. Aggression Taxonomy A pioneer in aggression research, Buss (1961) developed a taxonom y to categorize various types of aggression that served as a guiding structure for subsequent generations of aggression researchers. When considering aggression, the mo st salient examples include beating, shooting, shoving, stabbing, etc. In fact, the majority of psychological studies ha ve assessed physical forms of aggression by such measures as willingn ess to shock other participants or blast them with annoying and painfully loud noise. However, aggression also include s behavior such as verbal assault, rumor spreading, and ostracism. To capture the va st array of aggressive actions, Buss proposed three dimensions: physical vs. verbal, direct vs. indirect, and active vs. passive. Physical versus Verbal Although physical aggression m ay be the most obvious example, Buss (1961) emphasized that physical acts are not the sole means by which individuals can be aggr essive. Verbal actions

PAGE 13

13 can harm as well, but verbal aggression causes ps ychological harm rather than physical pain and injury. Specifically, Buss proposes two types of psychological pain: threat of future violence or rejection with rejection incorporating any type of re sponse that labels the victim as aversive, bad, unwanted, or lacking in some way. Some common examples of verbal aggression as rejection include aversive teasing of an unpopular peer, publicly denouncing a persons character, fabricating lie s about another person, derogating anothe r person in an argument, etc. Active versus Passive Busss (1961) original taxonom y also include d the difference between active and passive aggression. Buss noted in his brie f mention of this dimension that aggression can vary in the degree to which it is actively vs. passively committed. Specifically, active aggression requires taking initiative to construct an impediment th at prevents the victim from satisfying a goal, taking action to hurt another, etc.; in other words, it requires some degree of effort on the part of the aggressor to deliver noxious stimuli to the vi ctim. In contrast, passive aggression does not require the aggressor to act and in fact the very fact that the aggressor is not acting may constitute or allow the aggressi on. In other words, the simple pr esence of the aggressor or the aggressors inaction may be aggressive. For example, failing to render help may be seen as passive aggression. Witnesses to a crime who fa il to aid a victim may be accused of passive aggression if the witnesses intend this to be the side-effect of their inaction. Using this definition, it may be possible to classify so me witnesses to the famous Kitty Genovese murder as passively aggressive, especially as her homosexuality may have given her neighbors sufficient aggressive motivation. Consider another exam plea passive aggre ssor may fail to move out of anothers path simply to irritate the victim. Passive aggr ession is likely to be a successful strategy for the person motivated to avoid repercussions. Buss (1 961) argues that Passive aggression is a subordinates best weapon agains t his superior. Active attack invites retaliation. When the attack

PAGE 14

14 is passive, however, it is usually difficult for the victim to establish blame or to determine whether aggression has occurred (p. 9). Unfortuna tely, social psychologists focused primarily on active aggression with little research attentio n to the more passive and subtle forms of aggression. Despite social psychologys neglect of passi ve aggression, research on passive aggression is quite popular among clinical and counseling psychologists. Res earch on passive aggression led to the classification of the pa ssive aggressive personality disorder (also known as the Negativistic Personality Type). Acco rding to this conceptualization, passive aggression is seen as a means of aggression for individuals restricted in their ability or willi ngness to confront another in a direct and open manner.1 The Millon Clinical Multiaxial I nventory-III (MCMI-III) described the passive aggressive personality type as an ambivalent orientat ion in which people are torn between following the rewards offered by others and those they themselves desire.These individuals experience endless wr angles and disappointments as th ey vacillate between deference and defiance, obedience and aggressive opposition. Th eir behavior is charact erized by an erratic pattern of explosive anger or stubbornness inte rmingled with periods of guilt and shame (Millon, Davis, & Millon, 1997, p. 18). Some of these behaviors match Busss (1961) original conception of passive aggression bu t not all. Specifically, Buss s uggests that individuals with a need to be covert in their aggression may c hoose passive aggression, and this is somewhat reflected in the ambivalent orientation of wa vering between deference and defiance, but the anger and stubbornness are unique to this concep tualization. The prim ary means of passive aggression (failing to act rather than actively ini tiating a disruptive behavior) clearly establishes 1An abundance of research on the topic le d researchers to abandon it as a useful classification because criteria for diagnosing the disorder were imprecise and hard to substantiate.

PAGE 15

15 difficulty in assigning blame for harm to the vict im; in determining cause for some event, people look for changes or actions deliber ately taken. Failure to act is a more difficult basis upon which to cast blame. For example, failing to help so meone could be a form of passive aggression and more difficult to pass blame for such a behavior versus casting blame for physically assault. The MCMI-III (Millon, Davis, & Millon, 1997) also character izes passive aggressive people as expressively resentful by resisting the fulfillment of others expectations, procrastinating, etc., and other behavior that receives gratif ication in the demoralizing and undermining of others pleasures and aspirations. These individuals are also thought to have a discontented self-image in that they see themselves as misunderstood, luckless, underappreciated, and demeaned by others. The implication from this depiction of the passive aggressive personality disorder is one of a pe rson who feels deserving of the appreciation of others but does not believe that he/she receives it to the degree deserved. In a sense, then, although they realize that they are devalued by others, they feel this devaluation is undeserved as juxtaposed with their own high-va lue self-assessment. Therefore, the passive aggressive person may be somewhat narcissistic. The characterization of the passive-aggressive individual as someone who believes that he/she is underappreciated and demeaned and may re sort to behaviors that thwart others implies two things. First, the characterization of some one who is demeaned im plies that the passive aggressive individual is in a role of equal or lower status to th ose demeaning him/her. Moreover, the ability of this person to thwart the goals of ot hers implies that he/she plays an active role in the lives of others. In short, individuals may be more likely to use passive aggressive when aggressing against those higher in status and those with whom one expects ongoing interaction. In contrast, aggressing against lo wer status individuals poses a lower need for covert aggression,

PAGE 16

16 and so more openly and overtly hostile acts may occu r to those lower in relative status than those higher in relative status. Direct versus Indirect The direct vs. indirect dim e nsion of Busss (1961) taxonomy describes whether or not the victim is present at the moment of aggression wh en harm is inflicted. Direct aggression is that which requires a face-to-face confront ation, i.e., aggression done in the presence of the victim. In contrast, indirect aggression is done in the victims absence. Frequently, direct aggression is misconstrued as physical aggression and indirect aggression as verbal aggression. Although the overlap may frequently be true (hand-to-hand comb at is considerably easier when in the victims presence), it is not always the case. For example, an aggressor may engage in indirect physical aggressive by adding poison to food that the victim will eat hours later. Similarly, verbal aggression can be both direct (s uch as teasing someone) or i ndirect (by spreading gossip). Several authors have noted that indirect aggression is a particularly shrewd method to pursue because it helps to prevent a counterattack: because the aggressor is not present at the moment of aggression, it is difficult to identify the aggressor thereby hi ndering retaliation (Buss, 1961; Bjrkqvist, sterman, & Lagerspe tz, 1994; Bjrkqvist, sterman, Lagerspetz, Landau, Caprara, & Fraczek, 2001). In fact, this behind-the-back na ture of indirect aggression was originally construed by researchers to mean that the perp etrator would remain a nonymous to the victim; however, in the strictest definiti on, indirect aggression does not neces sarily protect identity. This focus on anonymity has since been expanded to incl ude the fact that indi rect aggression may not hide the identity of the aggressor, but simply the aggressive motivation for a particular action (Archer & Coyne, 2005). Bjrkqvist et al. (2001, p. 113) argue that i ndirect aggression attempts to cause psychological, in rare cases even physical, harm to the target person by social manipulation, often [but not always] attacking th e target in circuitous ways through a third

PAGE 17

17 person to conceal the aggressive intent, or otherw ise pretending the attack was not aggressive at all. Special Note on Terminology: Indirect, Relational, and Social Aggression Covert aggression, by n ecessity, must leave l ittle clear evidence th at implicates the offender. This will likely lead individuals to employ means of aggressing that leave little physical evidence, are subject to ambiguity and lend themselves easily to alternative interpretations. Of the domains available to the aggressor, physical, dir ect, and active aggression are particularly likely to conve y a clear intention of aggressio n. Individuals may choose verbal, indirect, or passive aggression to aggress because tone-of-voice can be reinterpreted, eye-witness accounts are missing, and volitional actions are missing. As a result, the domain of behaviors likely to be chosen by the covert aggressor incl ude a set of behaviors (e.g., aggression committed through social networks with social implicati ons for the victimteasing, ostracism, gossip, sarcasm, etc.) that have been classified in multip le ways. Indirect aggression, in the purest sense of Busss original work (1961), does not imply intenti on to do social harm or the use of others to accomplish ones misdeeds. However, much of th e previous work on indirect aggression has come to be nearly synonymous with aggression of a social nature. Archer and Coyne (2005) identified three distinct terms that share similar characteristic s: indirect aggression, relational aggression, and social aggression. The term relational aggression was first introduced by Crick and Grotpeter (1995) in their study of aggression amongst preadolescent children. The aut hors later defined relational aggression as behaviors that harm others th rough damage (or the threat of damage) to relationships or feelings of accep tance, friendship or group inclusio n (Crick et al., 1999). In its purest sense, then, relational aggression can be either indirect (such as gossip) or direct (such as a scorned lover who confesses all th e torrid details to the ex-lovers spouse); hostile/emotional or

PAGE 18

18 instrumental (damaging a victims relationship or social standing in a group to create new social opportunities for the aggressor); or verbal. The term social aggression was originally coined by Cairns et al. (1989) and defined as the manipulation of group acceptance through alie nation, ostracism, or character defamation (p. 323) and later expanded as directed toward damaging anothe rs self-esteem, social status, or both, and may take such direct forms as ve rbal rejection, negative facial expression or body movement, or more indirect forms such as slan derous rumors or social exclusion (Galen & Underwood, 1997, p. 589). In contrast to relationa l aggression, social aggression focuses not on the means of aggression but rather on the intended harm. In other words, social relations are the goal, not the tool, of aggression. Archer and Coyne (2005) argue that these three terms are fairly synonymous and equally viable alternatives to physica l aggression and distinctions between the three terms causes unnecessary conceptual complicatio ns. Although the behaviors that characterize these three types of behaviors are very similarif not identical ignoring the differing motiv es behind the various types of aggression, as Archer a nd Coyne implicitly suggest, can have negative repercussions in trying to gain a thorough understa nding of these behaviors. Behaviors may seem similar at a surface level, but the reasons behind these beha viors can vary dramatically. However, more importantly, failing to consider self-presentati onal concerns and the aggressors desire to embrace both a nonaggressive identity and the efficacy of aggression neglects much of what the would-be aggressor must consider. Without prope rly understanding all of th e considerations in the mind of an aggressor, a full understanding of aggression in an ever yday context cannot be attained.

PAGE 19

19 Gender Differences In early aggression research, res earchers concluded that men were more aggressive than women (Baron & Richardson, 1994). However, upon closer examination, researchers have tended to since conclude that when gender di fferences occur, men and women tend to choose different aggressive strategies that suit their particular gender c onstraints such that males choose direct physical aggression and females choose indire ct verbal aggression (Bjrkqvist, Lagerspetz, Kaukiainen, 1992; Crick & Grot peter, 1995; Lagerspetz, Bj rkqvist, & Peltonen, 1988). Although there is a tendency for men and women to differentially employ certain aggressive strategies, the reason for this is subject to interpretation. One path of reasoning may point to biological differences between the sexes that lead women to pursue a course of aggression more suited to their weaker physical strength wher eas men have the physical prowess necessary to invoke direct physical aggression. However, another path of reasoning suggests that women employ indi rect aggression more than direct aggression because of differential social resource s. Specifically, women are more socially connected than men (Green, Richar dson, & Lago, 1996) and therefore face greater repercussions for direct aggre ssion; e.g., attacking a single indi vidual in a densely connected social network brings with it a greater likelihood that the victim has social connections that will come to the defense of the victim (versus atta cking a single individual in a less dense social network). At the same time that a dense social network inhibits the like lihood of direct physical attacks, it simultaneously provides the resources necessary to utilize indirect (or social aggression) because this frequently requires the complicity of others (Green, Richardson, & Lago, 1996). Because men are less socially conn ected, they have fewer consequences for utilizing direct aggression and fe wer resources to utilize indirect aggression. Green et al. (1996) propose that the gender differences in aggressi on type are merely an adaptation made to

PAGE 20

20 contemporary challenges of me n and womenwhen controlling for social-connectedness, men with denser social networks re ported more indirect than di rect aggression and less direct aggression than men with low-density social networks, thereby demonstrating that the key difference is social connectedness, not gender. In deed, as children progr ess into early adulthood, these developmental differences should decline; this is in fact the case for 23 year-olds (Bjrkqvist, Kaukiainen, Laurila, & Ylinen, 199 4; as cited in Green, Richardson, & Lago, 1996). Intelligence and Indirect Aggression Just as gender-typed social networks can im pact the type of aggression one chooses, other individual differences of a more dispositional nature also influence choice of aggressive tactics. Several theories contend that one key individual difference is intelligence. For example, Tedeschi and Felson (1994) argue d that individuals with greater intellectual capacity are less likely to resort to aggression of any sort (versus other tactics such as ne gotiation), but should also be less likely to resort to the more brutal forms such as physical aggression. In other words, strong intellectual abilities allow individuals to circumvent dir ect physical attacks by means of more refined verbal or indirect attacks on anothe r. This view elicits th e possibility that less intelligent individuals should show more overt aggression than their more intelligent and covertly aggressive peers. Similarly, to predict maladaptive social be havior and aggression, Crick and Dodge (1994) proposed an information processing model. This m odel outlines a series of steps that individuals (in particular, children) take wh en reacting to others. The social skills deficit model proposes a direct link between bullying behavior and a dearth of social skills. In other words, individuals who lack the necessary social skills to resolve conflict peacefully may re sort to aggression as their only recourse.

PAGE 21

21 The model outlines six distinct steps individuals take when acting in the social environment and these researchers argue that aggr essive maladjusted children exhibit specific differences from nonaggressive socially adjusted children. First, individuals must encode cues from their environment (i.e., notice what is happ ening). Second, individuals must interpret these cues and give them meaningwas the shove from a classmate accidental or intentional? Third, individuals must select a goal a desired end state, such as obtaining a toy, making a friend, getting even, or staying out of trouble. Fourth, individuals ge nerate a host of possible behavioral responses that may aid the attainment of the goa lbeing nice to the ot her person, sharing a toy with the other person, reporting a tr ansgression, etc. Fifth, individua ls must select one of these generated responses and sixt h, enact that response. In an extensive review of this model, the au thors report findings indi cating that aggressive children are more attentive to aggressive cues in the environment, more likely to interpret cues as aggressive, generate fewer respons e options, and are more likely to select an aggressive response option. The seeming resolution to this problem is to teach these children a greater range of response options, how to better select options with a positiv e outcome (e.g., teaching children that playing nicely with others is more likely to win friends th an punching their peers), and how to avoid misinterpreting envir onmental cues as aggressively motivated. In line with this reasoning, Bjrkqvist and colleagues (2001) describe anti-bullying programs that teach children a greater appreciation of their peers emotional responses to aggression as a means to better predict the outcome of their aggressive acti ons (presumably, rejection by ones peers). Although this model seems quite logical and ma tches the popular conceptualization of the slow-witted bully, it makes some critical ove rsights in understanding aggressive children. Specifically, Crick and Dodge ( 1994) investigated physical and direct forms of aggression by

PAGE 22

22 operationalizing aggression as the extent to which children start fights, hit, push, or threaten peers (p. 82). Moreover, this social informa tion processing model is an attempt to explain maladjusted behaviors broadly and the authors perceive aggression as so closely linked to social rejection that they frequently use the term s aggressive and maladjusted interchangeably. However, this specific operational definition of aggression limits the model to specific types of aggression: direct and physical. By failing to incorporate additional behaviors, this model does not capture the full array of aggressive behavior s because it overlooks the indirect and covert forms of aggression. These types of behaviors ma y not fit within the framework Crick and Doge promote of the socially inept aggressor. In keeping with this possibilit y, this model fails to account fo r data that find positive social outcomes for aggressive children (e.g., LaFontan a, & Cillessen, 2002; Salmivalli, Kaukiainen, & Lagerspetz, 2000). Specifically, these authors found that the us e of indirect aggression as reported by peers was positively related to peer acceptance (Salmivalli, et al., 2000) and that relational aggressors may even attain higher sociometric status than their less aggressive peers (LaFontana, & Cillessen, 2002). In other words, an aggressive response option seems to be associated with at least some positive outcomes and, depending on ones goals, may actually be a fairly adaptable response choice. These resultant positive outcomes may then increase the likelihood of using aggression, just as any other reward increase s behavior. Therefore, it seems that covert aggression requires a certain degree of social savoir faire; social illiteracy would sabotage attempts at indirect aggression. Indeed, other research ers such as Kaukiainen et al. (1999) even found that higher levels of empathy (the aspect of empathy that involves understanding the emotional perspectives of others) actually lead to higher levels of indirect

PAGE 23

23 aggression. This tendency may cause the anti-bully ing programs described above to backfire by giving the bullies new tools with which to cause harm. This seeming disparity between the social information processing model (Crick & Dodge, 1994) and these other authors work can be underst ood in terms of the specific aggressive types the authors are each considering. Indirect aggressors may or may not differ from the direct aggressors on which Crick and Dodge focused in terms of cue encoding, cue interpretation, or goal selection, but individuals c hoosing to aggress indirectly s eemingly possess greater insight into the psychological state of th eir victims, making their aggre ssion all that more powerful and harmful. Moreover, this greater insight lik ely also helps the children avoid the same repercussions direct aggression w ould elicit because as indirect, covert aggressors armed with increased understanding of their victims, th ey may know better whet her their victims will retaliate, confront, or report them. Finally, indirect, covert aggressors may also differ in the generation of response alternatives and selection of responses. Bjr kqvist et al ( 1992) argue that aggressive strategies change as children deve lop new skills; children first use physical aggression when they develop self-motor control, then verbal aggression when they master sentence construction and vocabulary, and then finally indirect/social aggression when children develop social relationships and social skills. Therefore, more socially developed children may generate very different response options to an aggressive situation than less socially deve loped children. In sum, children seem to select the respons e options that best su it their own abilities: socially maladjusted children select aggressive responses that do not rely on social skills or the social network because they do not possess these t ools whereas socially skilled children select aggressive responses that maximize their own capital: a social networ k and skilled social interaction. Finally, Crick and Dodge s position assumes that socia lly adjusted children perceive

PAGE 24

24 the nonaggressive response alternative (if they ge nerate such an alternative as a behavioral option) as more attractive and bett er suited to attaining their goal s. For at least some people, a nonaggressive response option may be seen as less effective, and selection of such an option would be a mistake. Self-Presentation and Reputatio n in Aggressions Aftermath In traditional aggres sion experiments, participants are given the opportunity to aggress against other participants or c onfederatesvirtual st rangers. While true that people do aggress against those they do not know (e .g., rape and murder victims picked at random, mob aggression, etc.), most violence does not occur between total strangers. For example, in 2005, the U.S. Bureau of Justice estimates that seven out of te n rape victims knew thei r attacker and in 2002, of solved murders, 75% of the victims knew their attacker (Bureau of Justice Statistics Crime Characteristics). Although there are certainly some aggressi on processes that would be unaffected by aggressing against known others, aggressing against others who one must face later introduces a new problem. When one is motiv ated to harm another but must face the person in the future, it is worth avoiding being perceived as an aggressive indivi dual. Being perceived as nonaggressive maintains positive relationships and the element of surprise for later aggression. This conundrum of desiring a nonaggressive imag e yet desiring aggressi ve behavior can be thought of in terms of the eff ect/danger ratio. Bjrokqvist and hi s colleagues (1994) state that, the effect/danger ratio is an expression of the subj ective estimation of th e likely consequences of an aggressive act. The aggr essor assesses the relation between a) the effect of the intended strategy, and b) the dangers involved, physical, ps ychological or social, for him/herself, and for people important to him/her. The objective is to fi nd a technique that will be effective and, at the same time, incur as little danger as possible (p. 28-29). In other words, when individuals aggress against known others or those that they may encounter again in the future, there is the

PAGE 25

25 added concern of prolonged repercussions for their actions and the da nger component of the effect/danger ratio is heightened. In traditional studies, th e fear of future repercussions is minimized because the victim is someone that participants have never met and with whom participants have no reason to expect a future encounter. Neglecting to consider this aspect of interpersonal aggression ignores the fundamental human concern to maintain a particular imageimages that rarely include being aggressive. Although there is a desi re not to be seen as aggressi ve (particularly by those whom people must regularly encounter), the occasional desire to aggress against known others still exists. Therefore, people face a conundrum: how does one aggress against others while maintaining a nonaggressive image in the eyes of others? One possibility is using aggressive means that heighten the effectiv eness of the danger to compensate for the added danger of future interactions with ones victim. Another possibili ty is to reduce the danger component through the use of skillful self-presentation tactics. Self-presentation is the tool that people use to ensure that th eir desired images are maintained (Schlenker, 1980; Schlenker, 2003). Th ese specific images may vary from situation to situation and may be communicated through a variety of techniques such as mannerisms, anecdotes told, manner of dress, manner of sp eak, props, etc. Although self-presentation does not necessitate deceitfulness, it certainl y can be deceitful (such as claiming an identity that in no way reflects a persons actual image). Self-presentatio n also incorporates exag gerations of naturally occurring tendencies such as being friendly, re served, generous, caring, articulate, etc., and omissions or underplaying other qualities. Any particular image desired to be presented is motivated by the specific social situation alt hough some images may be universally desired. All self-presentations are restricted by reality and the believability of any given image claim: a

PAGE 26

26 person asserting a particular ident ity can successfully do so only to the extent that the witnessing audience possesses no information c onflicting with this identity. Keeping ones audience in the dark can be established by the audience a) possessing no information bearing on this claim (neither supporting nor c onflicting with the desired image) or b) possessing information that is at least somewhat consistent with th e desired image. In an aggre ssive situation, a person faces a difficult problem because his/her aggressive ac tions and traditionally desired nonaggressive identity are inconsistent. In order to successfully portray the desired image, one must make this inconsistency disappear. Specifical ly, the person must consciously disguise his/her actions as nonaggressive by creating enough ambiguity th at the victim or other audience a) must draw a nonaggressive conclusion or b) at least consider a nonaggression interpretation a viable option. Hardin and Higgins (1996) argue that all actors in the social environment simultaneously negotiate a shared reality; together, actors provi de explanations and interpretations of their world, and reality is defined as the interpretati on that all actors mutua lly agree upon. Actors are then obligated to treat this negot iated truth as truth. However, as actors encounter new situations and motivations, they may desire to renegotiate reality or find it necessary to defend the current reality. For example, perhaps a particular person previously considered a suitable friend commits a transgression serious enough to warrant being r ecast as an unsuitable friendship partner, or a reality in which a particular woman is consider ed nonaggressive is threatened by her aggressive actions. As such, individuals may provide alterna tive realities (e.g., recasti ng the role of the now unsuitable friend) or bolster the believability of the current negotiated reality by providing alternative explanations of the womans behavior. The ability to manipulate the situation and the conclusions that can be drawn is a valuable skill.

PAGE 27

27 Teasing provides one example of covert ag gression. In many interp ersonal relationships, teasing is considered a sign of affection a nd a way to demonstrate and increase social cohesiveness, although an abundanc e of research treats teasing as an aversive experience contributing to long-term maladjustment (e .g., Kowalski, Howerton, & McKenzie, 2001). As Campos, Keltner, Beck, Gonzaga, and John (200 7) point out, Teases then, are ambiguous communications that convey both negative and affiliative intent.Given this ambiguity, the relational consequences of a tease depend on the extent to which the target interprets the tease as an affiliative gesture that is an affront to the se lf (p. 4). In personal relationships characterized by teasing, the interpersonal partners may attr ibute any seemingly adve rse teasing as benign. Indeed, research supports the idea th at satisfied romantic partners wi ll adjust attributions of their partners to maintain a positive view of the part ner and relationshipexcusing negative behavior for those they like (Bradbury & Fincham, 1992; Goldstein, Tisak, Persson, & Boxer, 2006). However, this tendency may be abused when a relationship partner becomes upset with his/her significant other (or with ones frie nd) and lashes out verbally at ones partner. If confronted for this hurtful behavior, an aggre ssor can dismiss the aggressive e xplanation by claiming that he/she was only teasing and rely on the actual am biguity of teasing and the general positive attributions in the relationship to cover ones true intentions. Teasing is only one of several methods that individuals may use to aggress that easily lend themselves to alternative explanations in certain situations. For ex ample, aggressors who choose to ostracize a particular person can easily deny that the ostracism is intentional but rather merely accidental (the aggressors have plans with others, the vict im is paranoid, etc.). Similarly, gossiping about a particular person can be framed as venting without ma licious intent or as seeking advice. The use of double entendres may al so provide a safe route to express negativity

PAGE 28

28 that provides two plausible meaningsone innocuous meaning and another more risqu, offensive, or combative meaning. Other methods may include failing to aid a friend when aid is expected. For example, just as people try to manage the impressions that they themselves make on others, people also tend to help their friends make the impressions that will serve their friends best interests, thereby engaging in beneficial impression management (Schlenker & Britt, 1999). Consider a person who desires to impress a potential date or employer. He/she may ask a mutual friend to help create an image in th e eyes of the potential da te as an attractive and fun individual or as a competent and intelligent individual to the potential employer. The mutual friends failure to do so may be pe rceived as an act of aggression. The primary emphasis in social psychology is how individuals behave and respond to a social situation, frequently resulting in the characterization of peopl e as responding to the situation around them. This approa ch possesses a great deal of explanatory power but even more explanatory power can be gained by increased attention to the volition and decisiveness of individuals. Some research ha s demonstrated that people can and do actively construct their situation to influence the attri butions that audiences draw about them and their behavior. One particularly well-known example is self-handi capping through which individuals negate unflattering dispositional attribut ions in the event of failure outcomes while simultaneously augmenting flattering dispositional attributions in the event of successful outcomes (Berglas & Jones, 1978). The construction of an actual perf ormance impediment provides an alternative explanation for any forthcoming failure. This type of behavior indicat es that people are clearly se nsitive to the attributional tendencies of others and will actively try to control these attributions. Although constructing their own alternative attributions (via excuses, self-handicaps lies, etc.) may be preferred,

PAGE 29

29 individuals can also sophisticatedly capitalize on existing situations to ensure the most favorable attribution. One classic example of this ability is illustrated by Snyder, Kleck, Strenta, and Mentzer (1979) in which participan ts tried to control a ttributions that might suggest that they possessed negative attitudes toward physica lly handicapped individuals. Specifically, participants were told to sele ct a seat to watch a movie, choosing between watching a movie while sitting beside a person with a visible ha ndicap (wearing a leg brace) or watching a movie in a different room while sitting besi de a person without the visible handicap. In one condition, the persons handicap was th e only difference between the two rooms the same movie was playing in the two separa te rooms. However, in another condition, the rooms differed both in the person already seated and the movie play ing. Therefore, in the former condition, participants choosing one room over th e other might reasonably be perceived as choosing one room over the other based solely on the presence or absenc e of a handicap of the person already watching the movie. As such, choosing the room without the handicapped person would likely be considered an attempt to av oid the handicapped person. In the latter condition, however, choosing to watch the movie without th e handicapped person could be attributed to either avoiding the handicapped person or a desire to watch the differe nt movie in the other room (because both movie and confederate differed between the two rooms). Participants almost exclusively sat with the handicapped confeder ate when there was no plausible alternative explanation (same movie showing in both rooms). However, when both the movies and the confederates differed between the two rooms (provi ding attributional ambigu ity), participants did not seek out the handicapped i ndividualchoosing to sit with the non-handicapped person could easily be attributed to the movie selection rather than disliking the handicapped person. In other

PAGE 30

30 words, participants were aware of the conclu sions that audiences w ould draw about their intentions based upon their actions and f actors present in the environment. In a similar manner, Gaertner and Dovidio ( 1977) demonstrated that white participants were less likely to help black participants if failin g to help could be attributed to something other than their own prejudice. In this study, white participants did not differ in their willingness to help white versus black confederates if particip ants believed that they were the only ones who could offer help. In contrast, white participants were significantly less likely to help black confederates when they believed that other people could offer help. In other words, when white participants have only themselves to blame for leaving a confederate unaided, they offered help regardless of race. However, if participants failur e to help could just as easily be attributed to another reason (i.e., others could he lp the person in need), then participants were significantly less likely to aid the black confederate than the wh ite confederate. In short, the participants acted in accord with their socially proscribed prej udice when they believed it would not be easily detected, but restrained from acting on this prej udice if their actions clearly identified their prejudice beliefs. Together, these two studies (Gaertner & Dovidio, 1977; Snyder, Kleck, Strenta, & Mentzer, 1979) along with a host of additional research suggest that individuals are clearly aware of the attributions that others ar e likely to make and are motivated to control these attributions to preserve a desirable image in the eyes of others. This awareness of the a ttributional tendencies of others may curb the use of aggressi on in situations that enhance c onclusions that th e individual is acting in a deliberately hostile manner. Instrumental aggression in particular, even when enacted covertly, may lead observers to deduce that th e individual is acting in a deliberately hostile manner rather than in a nondeliberate manner. Hos tile/emotional aggression is carried out purely

PAGE 31

31 for its own sakeaggression that is instigated by emotion such as anger, frustration, etc., and carries with it some personal gratif ication. In contrast, instrumental aggression is carried out as a means to some other end (be it good or bad in nature). Of course, as Bushman and Anderson (2001) point out, aggression may be multiply motivated: aggression may be instrumentally motivated, but still provide a measure of emotiona l satisfaction, if nothing more than a reduction in physiological arousal after aggr essing (Geen, Stonner, & Shope, 1975). As is frequently the case when classifying behavior, Bushman and Anderson (2001) argue that emotional vs. instrumental aggression is best characterized as a continuum from exclusively instrumental to exclusively hostile/emotional rather than a dichotomy. Likewise, covert aggression may serve multiple purposes. In keepi ng with the notion that social actors knowingly manipulate the attributions that others make for th eir behaviors, social actors may be especially guarded in their aggressive in terpersonal actions when their victims or other observers are particularly likely to interpret behavior as self ish. In other words, when a person clearly gains from an action (as is the case in instrumental aggression), others may be especially suspect of this person. In short, instrumental covert aggres sion may be more desirous but will be less likely than emotional covert aggression, to the extent that the persons gained rewards are evident to others. It is my contention that this type of behavior occurs re gularly amongst those motivated to aggress against others. Individuals weigh the risks of getting away with such behavior against the cost of acting on their proscr ibed desires to inflict harm respond aggressively, etc. Situational Factors Several factors can sys tematically affect an analysis of the cost-benefit ratio of social behavior. First, some groups and cultures proscribe openly expressing grievances more than others. For example, Cohen and colleagues demonstrat ed that U.S. southerners are more likely to suppress external indicators of anger longer than U.S. norther ners (Cohen, Vandello, Puente, &

PAGE 32

32 Rantilla, 1999). Some groups value openly expre ssing grievances as a testimony to honesty and genuineness, and view denials of animosity a nd covert aggression as fake, hypocritical and dangerous to the social fabric because indivi duals do not possess an accurate understanding of which individuals they can truly trust. In cont rast, other groups and cult ures have established norms that stress playing nicely, maintaining a f aade of pleasantry and civility and perceive open expressions of animosity as threatening, rude, uncivil, and/or immature (Cohen et al., 1999). Although groups that permit and encourag e the expression of grievances and anger toward others may show higher absolute levels of clear aggression and nastiness, there is no necessary reason to assume that the aggressive and negative motives differ across these two types of groups. Presuming that such motives do not differ as a function of normative grievance expression, individuals with negative motives in th e civility culture must resort to more covert means of expressing their negativity. Beyond cultural proscriptions against expres sing grievances to others within ones culture/group, there may be both cultural and practical concerns when expressing grievances against specific others such as those higher in status or auth ority. Particularly authoritarian cultures may pressure group members from expressi ng hostility toward othe rs higher in authority while simultaneously increasing the likelihood that those of lower status will face repercussions of hostility. In addition to these cultural and norm ative concerns, however, there exists the very real possibility that those higher in status generally have greate r ability to extract revenge for expressed hostility/aggression than those lower in status. Therefore, if individuals are motivated to aggress against a higher status person, the mo re cautious choice would be to aggress covertly. Indeed, aggressing openly can be quite foolhardy, such as openly threatening ones employer.

PAGE 33

33 In addition to the status diffe rential between a potential aggr essor and his/her victim, there are other important factors that may mitigate th e path chosen when facing aggressive motives. For example, some relationships may be marked by a generally tension-free nature with the rare occurrence of a desire to aggre ss or act on hostility. However, othe r relationships may be fairly tempestuous but their long-term survival is desired or necessary. For example, romantic relationships may be particularly tumultuous bu t still be valued and so openly aggressive behaviors may be undesirable. On the other hand, the relationship may not be valued in its own right, but a minimal level of civility must be main tained as a means to some other end; e.g., an employee may not care for his/her boss but must maintain a minimally positive relationship with the boss in order to ensure continued livelihood. Bo th of these types of relationships should be marked by more covert than overt aggression. Hypotheses First and forem ost, future interaction is a cr itical antecedent of covert aggression. Without future interaction, self-presentational concerns are reduced. When people do not expect a future interaction with someone else, th ere is no need to hedge bets by covertly aggressing. Instead, any desire to aggress will be expr essed overtly or stifled. Conditions in which participants do not anticipate a future interaction where their protagonist/victim pa rallel traditional aggression research and so little would be gained from investigating th is possibility. In strategic covert aggre ssion, a person must carefully ba lance two primary concerns: a desire to effectively harm and also a desire to maintain a no naggressive image. Maintaining a falsely pleasant faade encourages recipients of th ese covertly aggressive actions to perceive the entirety of the communication as pleasant rather than unpleasant, thereby serving a dual purpose. First, if the recipient of a cove rtly aggressive act is suspicious that the action is aggression, the ambiguous nature ensures a reasonable chance that the perpetrator will ev ade retribution for a

PAGE 34

34 crime that was doubtfully committed in the first place. Second, a person may act in a covertly aggressive manner in response to a potential a ffront. If one later lear ns that no affront was actually intended and the act was misperceived, on e has a ready alternative explanation for ones covertly aggressive behavior. In short, covert aggression serves a person well when he/she is motivated to do harm to another person but preven t detection, and covert aggression also serves as a strategic response to ambiguous communication from others. Responding in an ambiguous manner allows individuals the opport unity to play both sides of the fence until more information is gathered about the social si tuation. Until more information is available, a dually aggressive and pleasant communication allows both the main tenance of positive interaction and negative interaction simultaneously. Once more information is gathered, individuals can bolster the appropriate interpretation (pleasant or unpleas ant) through less ambiguous communication. Thus, covert aggression may initiate an unpleasant interaction or it may be used a cautious response to the ambiguous behavior of others. Because of the subtle duality of a cove rtly aggressive communication and based on previous research regarding indire ct aggression (a similar but dis tinct concept), one would expect that socially savvy/sensitive indi viduals would be particularly likely to employ this strategy both of their own volition and also in response to communications from others. However, when people expect no future interaction with someone, there is no n eed for covert communication or hedging bets because there exist few (if any) reperc ussions for overt aggression. In situations that elevate consequences for acting overtly aggressively, individuals should be more likely to choose covert aggression over overt aggres sion. This leads to Hypothesis I: Hypothesis I: The likelihood of covert aggression is heightened when future interaction is expected, particularly when one is partially or wholly dependent on the other in that future interaction.

PAGE 35

35 Moreover, the presence of a c overt aggressive path shoul d heighten the likelihood of aggressing than when individuals only recourse is open and direct aggression. This leads to Hypothesis II: Hypothesis II: Overall aggression (combined straight forward and ambiguous aggression) will be less when one is forced to aggress openly than when the opportunity to aggress covertly is also available. If no covertly aggressive opti on is available, individuals pa rticularly concerned about the consequences for aggressing at all will be le ss likely to aggress in any mannerovertly or covertly. The consequences for aggressing at al l are heightened to the extent that future interaction is likely or imminent, and to the exte nt that there are str ong proscriptions against hostile behavior. Such proscriptions can arise due to cultural climat e or status differentials (i.e., aggressing against individua ls higher in power). Specific types of individuals may be especia lly likely to engage in covert aggression. Because covert aggression and indirect aggressi on may involve the same types of behaviors, scales assessing the tendency to engage in indi rect aggression may also assess predilection to engage in covert aggression. Several different me asures of aggressive tendencies have been developed to assess predispositi ons toward aggression and aggre ssive personal histories. Of course, several scales focus on more overt form s of aggression, but some scales assess both proclivity to engage in direct overt forms of aggression as well as subscales tapping emotions that lead to the aggressive motive. For exampl e, the Buss-Perry Aggression Questionnaire (Buss & Perry, 1992) does not exactly assess indirect aggression, but it does include subscales assessing not only physical and verbal aggression but also hostil ity and anger. In particular, later researchers demonstrated that the hostility subscale predicted indirect ag gression better than the other subscale components and thus may be a usef ul tool in predicting whom is likely motivated to covertly aggress against others (Archer & Webb, 2006).

PAGE 36

36 Other scales have come closer to assessi ng covert aggression. Richardsons Conflict Response Questionnaire (Richardson & Green, 2003) is a 28-item scale that assesses the tendency for individuals to use various strategies under conflict. The two subscales consist of a score for both direct (e.g., frequency of pus hing, grabbing, or shoving someone; frequency of yelling or screaming at someone) and indirect aggression (frequency of spreading rumors about someone; frequency of telling others not to associate with someone). Other popular methods of assessing indirect aggression util ize peer nomination and/or ratings of all members of a specific group (frequently all students in a class) (Bjrkqvist et al., 1992). However, although this technique may work well with young children who have a limited and structured social world easily amenable to psychological research, this method is harder to employ with adults whose social worlds are not as easily compartmentalized into a si ngle classroom. In the current study, both Richardsons Conflict Response Questi onnaire and the Buss-Perry Aggression Questionnaire will be assessed as potentially moderating variables to predict likelihood of engaging in covert aggression and to test Hypothesis III and IV: Hypothesis III: Participants with greater proclivity toward (1) hostility, (2) anger, (3) verbal aggression, and (4) indire ct aggression will show highe r rates of covert aggression. Hypothesis IV: Participants with greater proclivity toward direct aggression will show higher rates of open aggression. It is possible that individuals low in Hos tility (as assessed by the Buss-Perry Aggression Questionnaire) or low in Indirect Aggression (as assessed by the Richardsons Conflict Response Questionnaire) may be less likely to (a) interpret an ambiguously aggressive response as aggressive and (b) employ ambiguous aggression in response. In addition to measures assessing aspects of aggression, increased empathy (particularly the perspective taking aspect of empathy) ac tually provides individuals with increased knowledge of their victim and ability to cause harm to their victims. Therefore, the perspective

PAGE 37

37 taking and empathic concern subscales of Davis s Empathy Scale (1980) s hould also be assessed to assess Hypothesis V: Hypothesis V: Individuals high in (1) empathy, especi ally(2) perspe ctive taking, will show increased likelihood of usi ng covert aggression. Finally, despite the covert meas ures by which participants will be able to aggress in the current study, some participants may be particular ly reluctant to act in a harmful manner, despite the temptation. Specifically, partic ipants high in need for appr oval should be less likely to engage in aggression of any sort, including covert aggression. To account for this possibility, the Crowne-Marlowe measure of Social Desirability (Crowne & Marlowe, 1960) will be included in the current study. Thus, Hypothesis VI: Hypothesis VI: (1) Individuals high in social desirabi lity will be less likely to engage in aggression of any sort (either covert or ope n), but when they do, (2) they should show a greater preference for covert aggression than their low social desira bility counterparts. In short, the following predictions are made about each conditions aggressive behavior: Hypothesis VII: (1) Individuals receiving positive f eedback will respond in kind to their evaluatorselecting positive messages. (2) The level of positive and negative messages sent will not differ between positive feedback participants who have a covert method of aggression and those who do not have a c overt method of aggression available. Hypothesis VIII: (1) For participants expecting an interactio n, individuals receiving straightforward negative feedback who do not have a covert method of aggression available to them will send primarily neutra l messages and few negative messages. (2) For the negative feedback participants expecti ng an interaction but who do have a covert method of aggression will choose to send primarily double entendre messages (covert aggression), and few negative messages. (3) Nega tivefeedback participants with a covert strategy available to them but who are not expecting an interaction will send primarily negative messages and few (if any) doubl e entendres (cover t aggression). Hypothesis IX: (1) In general, participants who r eceive double entendre feedback (covert aggression), will use more double entendres th an participants who receive positive or negative feedback. (2) More specifically, the participants who expect an interaction will use more covert aggression than the participan ts not expecting an in teraction, who will use primarily negative statements with few doubl e entendres.(3) Of those expecting an interaction, participants with covert aggr ession available to them will use many double entendres and few if any negative messages whereas (4) participants without covert aggression available to them will send pr imarily neutral and positive statements.

PAGE 38

38 Moreover, in addition to actual behavior and messa ges sent to others (i.e., the actual act of aggression), beliefs about the othe r person will also likely be affected. These beliefs are unlikely to be affected by the response options availabl e to the participants. As such, the following predictions are made regarding the be liefs about the original communicator. Hypothesis X: Communicators sending positive feedback will be viewed most favorably by the participants in terms of (1) being liked the most and (2) participants expecting to be liked by the communicator. Little difference is expected on these dimensions between the communicator sending negative feedback and th e communicator sendi ng covert aggression (double entendre) feedback. Above and beyond the liking dimension and warmth of feelings toward the original communicator, the participants should differ in their estimation of the communicators communication skills and deserved respect. Alth ough being a recipient of covert aggression is anticipated to be an unpleasant experience and should manifest into dislike for the communicator, there may be admiration and respect for the skills necessary to successfully implement such a covertly aggressive maneuve r. Therefore, the following is predicted: Hypothesis XI: Communicators sending double entendre feedback will be seen as more skillful communicators than both (1) negativ e feedback communicators and (2) positive feedback communicators. (3) Positive feedback communicator s should be perceived as no more skillful than the negativ e feedback communicators, but a halo effect may lead to a slightly higher evaluation of the positive feedback communicators skillfulness than the negative feedback communicators skillfulness. As suggested by research on the self-serving bias, individuals who receive negative and ambiguous feedback should diminish the validity of the test to downplay the relevance of the test in order to minimize the self-imp lications of negative feedback. As such, recipients of negative and double entendre feedback should not internali ze the feedback they receive regarding their own creative intelligence. Hypothesis XII: Participants receiving (1) negative and (2) double entendre feedback will view the test as a less valid test of creativ e intelligence than the pa rticipants who receive positive feedback.

PAGE 39

39 Hypothesis XIII: Participants receiving negative a nd double entendre feedback should not internalize the negative fee dback, although negative feedback participants will be more likely to internalize th e feedback than ambiguous feedback participants due to the possible positive interpretation in the ambiguous feedback. The emotional state of participan ts should also vary as a result of the feedback that they receive. Specifically, emotions should map onto the interpreta tions of the feedback. These emotions and anxiety should motivate participan ts to escape face-to-face interaction when possible. Therefore: Hypothesis XIV: (1) Participants recei ving positive feedback should report positive emotions, (2) participants receiving negativ e feedback should report negative emotions, and (3) participants receiving double entendr e feedback should report both positive and negative emotions. Hypothesis XV: Participants receiving negative and double entendre feedback will be most likely to avoid face-to-f ace interactions and instead seek out less immediate forms of communication (i.e., computer mediated interaction and intercom mediated interaction). In sum, then, it was expected, generally, that participants would respond in kind to the feedback that they received. In other words, positive feedb ack participants would respond primarily in a positive manner, negative fee dback participants would respond in a negative manner, and ambiguous feedback participants would respond via mixed messages when available, and would eschew outright negativ ity for positive messages when mixed messages were not available (and only neutral messages were). Likewise, the ability to employ mixed messages should similarly curtail the use of negative feedback participants endorsement of outright negative messages. Moreover, a future in teraction was expected to restrain the use outright negative messages in favor of mixed messages. It was also expected that ambiguous feedback and negative feedback participants wo uld view their partners in a similar negative manner relative to positive feedb ack participants, who were exp ected to view their partners positively.

PAGE 40

40 Current Study The current study takes a unique approach to understanding interpersonal covert aggression. Little psych ological at tention has been given to the f acet of aggression that requires continued interpersonal contact a nd thus a need to maintain an interpersonal identity as a nonaggressive person. I contend that when individua ls anticipate future interaction with another person, the need for a costs/benefits analysis is heightened because in ongoing relationships, it is even more important to maintain a nonaggressive interpersonal iden tity. As such, people will be strongly motivated to avoid actions that will cause others to view them as aggressive rather than nonaggressive. However, because the motivation to aggress can be particularly strong, people will create opportunities or act on existing opportunities in the environment that capitalize on attributional ambiguity to escape dispositional a ggressive attributions. When another person is conveying potentially aggressive or demeaning intent, the recipient of this message must determine the true meaning behind the ambiguity and respond appropriately. If the message is misinterpreted as aggressive or negatively toned and one responds in a defensive manner or in a retaliatory manner, one unnecessari ly escalates an interaction in to an openly aggressive and unpleasant situation that could have deleterious effects fo r the ongoing relationship between the interactants. In contrast, if one incorrectly misinterprets the co mmunication as innocent, one may face the consequences of trusting those who shoul d not be trusted, failing to prepare against future less covertly aggressive in teractions, be perceived as foo lish and nave, etc. Therefore, when given the opportunity to rely on attri butional ambiguity (e.g., through the use of double entendres), people should be more likely to demonstrate aggressiveness than when no such attributional ambiguity exists. The extant research on aggression lacks experi mental investigation of the process by which individuals decide to engage in covert aggression, thei r success in using it, and the repercussions

PAGE 41

41 for themselves and others. In addition, although some research suggests that individuals who utilize indirect aggression (and possibly more speci fically covert aggression) are more successful interpersonally than those who do not use this strategy, the reason is unclear. The reason indirect aggressors (and possibly cove rt aggressors) tend to ha ve higher social status could be that a) the use of covert aggression actually increases social status by inspiring respect in others, or b) socially successful individuals are more ad ept at covert aggressi on but employing covert aggression does not actively influence ones social standing. The current study is an experimental analysis of this question.

PAGE 42

42 CHAPTER 2 METHOD Overview The curren t study assessed reactions to varying types of communication: positive, negative, and ambiguous evaluations with poten tially strong positive or negative connotations. These evaluations came in the guise of evaluative feedback regarding participants attempts to complete a creative intelligence test Most participants (with the ex ception of those in two offset conditions) expected a future interaction with th e other participant who provided the evaluative feedback. Participants then had the opportunity to send a retu rn communication to the other participant in response to the evaluative feedback Some participants were then provided with a way to aggress covertly whereas others did not have this path available to them, and instead were only able to respond in a directly positive, negative, or neutral manner. Participants also reported their perceptions of the other person to the expe rimenter and then selected what method of communication (face-to-face, speaker system or computer mediated communication) they wished to use for the secondary group-ba sed creative intelligence assessment. The communication method served as measure of ho w much participants chose to distance themselves from the other participant. Design The design was a 3 (Feedback: explicitly pos itive, explicitly negative or am biguous overture) X 2 strategy (Strategy: mixed messages vs. neutral message res ponse option) factorial. Two offset groups that did not expect any interaction underwent the critical manipulation receiving either negative or ambiguous feedback with an opportunity to respond via mixed messages. These offset conditions were incl uded, pending variations in the use of mixed messages, to test the importance of anticipating an upcoming interaction.

PAGE 43

43 Participants Participants (N = 295) who reported English as their prim ary language were recruited from introductory psychology courses. Of these, 68 pa rticipants (54 females and 14 males) were randomly assigned to the offset comparison condi tions leaving 227 particip ants in the main study design (170 females and 57 males). Participants we re given credit toward completion of their course research requirement. Prescreening Session Prior to arriving at the m ain experimental session, participants participated in a brief prescreening session to gather i ndividual difference information. As such, they completed the Richardson Conflict Response Questionnaire (Richardson & Green, 2003), the Buss-Perry aggression Questionnaire (1992), a nd the Social Desirability Scale (Crowne & Marlowe, 1960). To enhance the cover story, participants al so completed some filler questions tapping extraversion, conscientiousness, and openness to experience. See Appendix A. Procedure Participants were run individually. When part icipants arrived, they were greeted by an experimenter, seated in a room by themselves, and asked to read and sign an informed consent, read a Study Overview briefly describing the purpos e of the study. Participants were told that the study was investigating group problem solving versus individual problem solving efficacy throughout early adulthood. As su ch, participants first completed a creative problem solving assessment by themselves that ostensibly served as a comparison baseline to a second creative problem solving assessment completed collaborati vely with another person. Prior to completing the first assessment, participants completed the perspective taking and empathic concern subscales of the Davis Empathy Scale (Davis, 19 80). Participants were also asked to briefly

PAGE 44

44 describe why they were attending the University of Florida, their future plans, and what they enjoy doing for fun on the Basic Inform ation Questionnaire. See Appendix B. Participants were told that the individua l version of their creative problem solving assessment asked them to design a squirrel-proof bi rdfeeder and that their fellow worker was in a separate room working on a different form of the same test (a moon crash landing survival problem). To do so, participants were given six minutes to design, diagram, and explain a birdfeeder design that would prevent squirrels from accessing birdseed but still allowed birds access. Participants were given several sheets of blank paper to complete their design. After completing the first assessment, the experimenter returned and collected the participants designs. The partic ipants were reminded that in many real-life work situations, people frequently receive feedback from co-worke rs that can be helpful when later tackling a similar problem and that such information can be elucidating. The curre nt study was attempting to mimic that tendency in the real world by allowing the two individu als to evaluate and comment on each others initial assessment prio r to working together on second assessment. Participants were presented with the results of the other participants (t he fellow worker) creative problem solving assessment. Again, under the gu ise of mimicking real-world interaction in which individuals must provide evaluations of th eir peers work, participants were asked to evaluate the quality of the other participants so lution to his/her problem via a free response item. The other participants solution to the problem was a reasonabl y good but imperfect solution to the other participants own problem. While the participant was evaluating the fellow workers solution, the fellow worker was supposedly evalua ting the participants ow n drawing, and after a few minutes, the experimenter retu rned to retrieve the participan ts evaluation before exchanging them with the evaluation of the participants own drawing by the fellow worker and the fellow

PAGE 45

45 workers Basic Information Questionnaire (see Appendix C). The Basic Information Questionnaire conveyed a basic level of informa tion about the fellow worker and indicated the other participant was the same gender as the participant himself/herself. Participants anticipating an interaction were randomly assigned to either the positive, negative, or ambiguous feedback condition. The pos itive feedback indicated the fellow worker perceived the participants desi gns as generally good, as showi ng a good deal of potential, and praising the participants creative intelligence. Th e negative feedback indicated that the fellow worker perceived the participants designs as faulty and uncreative, and disparaged the participants creative intell igence. The ambiguous feedback conveyed two possible interpretations to the participants: both a poten tially positive meaning and a potentially negative meaning. The ambiguous feedback was pilot tested to determine that, give n the nature of the task, participants perceived the feedback to c onvey these two potential meanings. (See Appendix D). Participants assigned to one of the two offset conditions not expecting an interaction were assigned to receive either negative or ambiguous f eedback, but not positive feedback as that was not a primary concern for the current study. After reviewing the Basic Information Questi onnaire and reviewing the fellow workers feedback, the participants were then asked to communicate their in itial impressions of the person to both the fellow worker himself/herself via the First Impressions Questionnaire and subsequently to the experimenter via the Preliminary Interaction Assessment. The First Impressions Questionnaire was supposedly sent to the fellow worker to fa cilitate the secondary assessment in that it is helpful to know what other people believe about oneself. On this questionnaire, participants were presented with a list of eighteen statements. Half of the participants were randomly assigne d to receive a checklist that c ontained three different types of

PAGE 46

46 descriptors, positive, negative, and mixed me ssage descriptors (conveying both a positive and negative connotation). For example, a positive de scriptor read, This pe rson is capable of doing many things. A negative descri ptor read, This person finds novel tasks difficult. A mixed message descriptor read, The quality of th is persons work is well-known. (Appendix E.) The other half of the participants were random ly assigned to receive a checklist that only contains positive, negative, and neutral descript ors (no double entendres). (Appendix F.) In short, the participants in this cond ition had no ambiguous path availa ble to them by which to respond safelythey had to choose to respond in a directly positive, negative, or neutral manner. Selecting neutral descriptors allo wed individuals to take a middl e-of-the-road approach but did not allow them the option of presenting a dually friendly and aggressive image. A neutral approach presented neither of these images. The eighteen statements were equally divided across the three types of responses presen ted to the participants (positive, negative and either neutral or mixed messages). Participants were asked to check off up to 8 items that they believed accurately described the person with whom they were interac ting. After selecting 8 items, participants were asked to rank these 8 statements from most desc riptive to least descrip tive of the other person. This approach allowed participants to select all or none of the items in each categoryif participants wished to eschew double ente ndres entirely, they had that ability. After participants sent this information, they we re told that the next phase of the study was the collaborative second assessment. Under the gu ise of assessing any extr aneous variables that could potentially affect performance on the colla borative assessment, participants completed the Positive and Negative Affect Scale (Watson & Tellegen, 1988) described as part of a Preliminary Interaction Assessment (see Append ix G), as emotional responses may have mediated the extent to which participants derogate the fell ow workers personality or ability.

PAGE 47

47 The remainder of the Preliminary Interaction Assessment assessed the feelings that participants had toward the other person, the im pending interaction, and themselves such as how much they anticipated liking the other person, anticipated being liked by the other person, assessment of the others social skills, etc. This questionnaire gave the participants the opportunity to express their nega tivity toward the fellow worker more openly because they were told that this questionnaire w ould not be shown to the fellow wo rker. Moreover, it was stressed that it was important to answer honestly to ensure validity of the final assessment. In addition to the previously mentioned questions, this questio nnaire also assessed how positively participants believed they evaluated the other participants initial assessment. I predicted that because participants woul d not have received feedback on their own performance when evaluating the performance of ot hers, they would have no reason to assess the performance of the other in a strongly negativ e way. Previous research shows that when evaluating a specific other person, people tend to evaluate that pe rson fairly positively (as above average; Klar & Giladi, 1997). As such, participan ts evaluations of the fellow workers initial creative intelligence assessment would be fairly po sitive. The participants evaluations of the partners work were objectively assessed by inde pendent raters at the conclusion of the study. However, participants who received a negative or ambiguously aggressive evaluation may have altered their retroactive evaluations of the ev aluations they gave ear lier by doing one of two things. First, they may have inflat ed the positivity of their evaluati ons to bolster their justification for a hostile reaction or, alternativ ely, they may have deflated th e positivity of their evaluations to portray themselves as less nave in their willin gness to play nice; in short, participants may have portrayed themselves as having more of an upper hand than they actually had. Questions

PAGE 48

48 assessing this possibility were onl y exploratory in nature, as either of these two possibilities were equally viable possibilities. Finally, the experimenter told participants th at there was a degree of flexibility regarding what communication method the participants used to collaborate on the final assessment and because the study had not been active for ve ry long, the study still needed people to communicate in all methods. Therefore, the experi menter offered the participant a choice. The participant could choose between interacting face-to-f ace with the fellow worker, via a speaker system (only audio transmission), or via a computer chatting program. This measure assessed the degree to which participants desired control over their comm unications and psychological distance from the communication with the fellow worker. After the participants made their selection, the partic ipants were carefully probed for suspicion, debriefed, and dismissed.

PAGE 49

49 CHAPTER 3 RESULTS Pilot Testing Both the feedback and th e statements in th e first impressions questionnaire were pilot tested to ensure that multiple meanings were able to be perceived. Several variations of ambiguous feedback were tested, as were several variations of neutral and mixed messages. For the ease of presentation, only the final se lected measures will be discussed. Feedback Participan ts (N = 77) completed the squirrel proof birdfeeder problem described above. Then, participants were asked to read one of the three possible versions of the feedback and asked to (1) report their interpretation of the fee dback, and (2) report, if such feedback had been actually based on their birdfeeder solution, how th ey would feel about re ceiving such feedback. Specifically, participants reported how positiv e an evaluation the feedback conveyed, how negative an interpretation it conveyed, and how mu ch a negative and positive interpretation were simultaneously conveyed. Moreover, participants reported how sincere the author of such feedback was, how upsetting the feedback would be and how much the author of such feedback was trying (1) to conceal an insu lt and (2) to be superior. All of these ratings were made on a scale of 1-9 Likert-type scale. See Appendix D for the specific wording of the feedback and dependent measures. A significant effect of feedback wa s found for ratings of positivity ( F (2, 74) = 126.39, p < .0001, p 2 = .774), negativity ( F (2, 74) = 111.58, p < .0001, p 2 = .751), and simultaneous positive and negative meanings ( F (2, 74) = 8.22, p = .0006, p 2 = .182). See Table 3-1 for mean ratings as a function of feedback. Specifical ly, although participants perceived the ambiguous feedback to be as positive as the positive feedback, they percei ved the ambiguous feedback to be

PAGE 50

50 more negative than the positive feedback. In ad dition, the ambiguous feedback was perceived to convey a double-meaning significantly more than both the positive and negative feedbacks. Regarding visceral reactions and perceptions of the feedbacks au thors, participants expected being most upset after receiving the negative feedback (significantly more than both the positive and ambiguous feedbacks), and ambiguous feedback was seen as more upsetting than positive feedback. In keeping with this pattern, pa rticipants reported that the negative feedback and ambiguous feedback authors were most trying to appear superior (sig nificantly more so than the positive feedback author). Ambiguous feedb ack was seen as trying to conceal an insult significantly more so than positive feedback; negative feedback was seen as intermediate and not significantly different from either positive or am biguous feedback. Likewise, the same pattern on the inverse question was evident such that ambiguous feedback was seen as significantly less sincere than positive feedback, while negative feedback was intermediate and not significantly different from either. In sum, ambiguous feedback was perceived as conveying both a negative and positive meaning, was more negative than positive fee dback and was more positive than negative feedback. Likewise, ambiguous feedback was mo re upsetting and less sincere than positive feedback, and was seen as more of an attempt to be superior and conceal an insult than positive feedback. See Table 3-1 for mean ratings of positive, negative, and double-meaning interpretation of each feedback as well as pilot participants reactions to the feedback and the feedback communicator. In short, the various forms feedback indica ted that they had the desired effect; the feedback was effectiv e in inducing the desired percepti ons in the pilo t participants. Message Options For the First Im pressions Questionnaire in which participants select and rank statements to send to their partner after receiving feedback, it was necessary to have statements with a single

PAGE 51

51 positive meaning, statements with a single negati ve meaning, statements with a simultaneously strong positive and a strong nega tive meaning (mixed messages), and statements that were relatively neutral i.e., a single low negative/slightly positive meaning. Th e same participants as above also rated possible statements as to how positive they were, how negative they were, and the extent to which they had both a positive and ne gative interpretation on a scale of 1 (not at all) to 9 (very much). The following guidelines were used to select statements for each category. Positive statements had to have a double-meaning score < 5, a positive score 6.5, and a negative score 3. Negative statements had to have a double-meaning score < 5, a positive score 3, and a negative score 6.5. Mixed message statements had to have a double-meaning score 5, a positive score 6, and a negative score 4. Neutral statements required a double-meaning score < 5, a positive score 6 but 3, and a negative score 4. See Appendices E and F for the final selected statements that met these crit eria. Thus, the messages from which the actual participants could select were rated by the pilot participants as having the desired properties. Scoring the First Impressions Questionnaire To determ ine the use of positive, negative, or alternate responses (mixed messages or neutral messages), the 18 statemen ts of the First Impressions Questionnaire were divided into the three categories: (1) 6 positive messages, (2) 6 negative messages, and (3) 6 alternate messages, which consisted of either mixed messages (for pa rticipants in the mixed message condition) or 6 neutral messages(for participants in the neutral condition). Participants selected 8 total messages and then ranked these 8 messages from most desc riptive (rank = 1) to least descriptive (rank = 8). Two types of information can be gathered from the First Impressions Questionnaire (FIQ). First, what messages did the participants se lect? For each category, the number of messages selected from each category was tallied such that if a participant selected 2 positive, 1 negative,

PAGE 52

52 and 5 alternative messages, then he/she would have a positive-selecti on score of 2, a negative selection-score of 1, and an alte rnative-selection score of 5. The second piece of information to be gathered from the FIQ is how descriptive of their partner the participants perceived the selected messages from each category to be (indicated by the rank-value assigned to each statement). Ther efore, rank-scores were computed by averaging the ranks assigned to the 6 messages from each category. Because ten messages were not selected and therefore not ranked, a rank valu e of 13.5 was assigned to these unselected 10 messages by computing the average value that wo uld have been assigned to the unselected 10 messages if all 18 messages had been ranked. For example, if a participant selected 2 positive messages and ranked them 1st and 2nd most descriptive and did not select the other four positive messages, then this participants positive -rank score is an average of 1, 2, 13.5, 13.5, 13.5, and 13.5. If none of the six messages from a given cate gory were selected, then the score for that category would be 13.5 as all messages in that ca tegory would be assigned a score of 13.5. If all messages from a given category were selected and ranked as most descriptive, then the score for that category would be 3.5 (i.e., an average of 1, 2, 3, 4, 5, and 6). Thus, the possible range of scores for each category range from 3.5 to 13.5 with lower scores reflected greater endorsement of that type of message. These two measures of category endorsement we re highly correlated w ith each other. As would be expected, as more statements were selected from each category, the average rank for each category was higher (where higher ranks are reflected by lower numerical values). Specifically, selection scores and ranking score were highly correlated for each category: r alternate statements (220) = -.91; r negative statements (220) = -.99; r positive statements (220) = -.98; all ps < .0001. Due to the extremely high correlation betw een these selection and ranking measures, only

PAGE 53

53 analyses regarding the average ranks assigne d to each category are presented for ease of presentation. The patterns of significant and non-significant findings were essentially identical for the ranking and selection measures, a nd any exceptions are indicated below. Overview of Analyses In the analyses below, partic ipan ts who anticipated an inte raction with their partner (N= 227) were included in a series of 3 (Feedback : Positive, Negative, or Ambiguous) X 2 (Response Option: Mixed Messages or Neutra l Messages) general linear models that tested the use of positive messages, negative messages, and alternative messages (mixed messages or neutral messages, depending on response option condition). Although the effects on each dependent measure are interesting in and of themselves testing specific hypothe ses requires comparing across categories. Because the three categories are not independent of one another, a degrees of freedom restriction occurs if all three categorie s are compared to each simultaneously. To test hypotheses that predicted differenc es across categories, difference scores were calculated to compare each of the three categor ies to the other two. These differences scores were then submitted to the same general linear model de scribed above. For ease of presentation, the statistical significance of mean comparisons desc ribed below are presented in Tables 2 and 4. Differences between specific conditions after a significant main effect involving more than 2 levels of a significant interaction were determined by tests of simple effects or specific contrasts. The results are indicated via subscripts in the tables. Gender was included as a possible moderator in each of the analys es, but failed to produce any significant interactions or main effects and so was dropped from the analyses. The offset comparison conditions were included in the current study to assess the extent to which any differences in the use of mixed messages were affected by the po ssibility of a future interaction or generalized to reactions regardless of future interactions. Therefore, the dependent

PAGE 54

54 measures of the offset comparis on conditions were assessed using the same statistical framework with follow-up comparisons describing whether the future interaction mattered. These analyses are described in more detail below. Messages Used Positive, Negative and Alternate Messages Variations in the averag e rank-value of each category of messages (positive, negative, and alternate) was determined for each combinati on of feedback and res ponse option using the 3 (Feedback) X 2 (Response Option) general linear m odel approach described above on each of the three message categories. The average ranks assigned to each category are presented in Table 32. Means for the average number of statements selected from each category (the selection scores described earlier) are presente d in Appendices H and I. In general, the hypotheses predicted that part icipants would respond to the feedback they received in a similar manner such that positive feedback participants would send positive messages, negative feedback participants would send mixed messages when they were available and negative messages when mixed messages we re not available, and ambiguous feedback participants would send primarily mixed messages wh en they were available but would resort to neutral and positive messages when mixed messages were not available. Further, it was predicted that the absence of mixed messa ges as a communication alternat ive would have no effect on the primarily positive messages sent by positive feedback participants. Regarding the use of alternate messages (neu tral or mixed messages), a significant main effect of feedback ( F (2, 214) = 12.46, p <.0001, p 2 = .104) and simple effects comparisons indicate that nega tive feedback participants e ndorsed the alternate messages ( M negative feedback = 8.54) more than either positive feedback particip ants and ambiguous feedback participants ( ps < .0001, M positive feedback = 9.65, M ambiguous feedback = 9.60). Surprisingly, ambiguous feedback

PAGE 55

55 participants and positive f eedback participants did not differ from one another ( p = .81). These findings are not consistent with the hypothesized conten tion that ambiguous feedback participants would most strongly endorse mixed messages as negative feedback participants were the most likely to endorse mixed messages/ neutral messages and positive and ambiguous feedback participants generally avoided them. Negative messages were also submitted to the above model. Main effects of both feedback ( F (2, 214) = 123.32, p < .0001, p 2 = .535) and option ( F (1, 214) = 6.18, p = .01, p 2 = .028) were qualified by a significant Feedback X Option interaction ( F (2, 214) = 7.23, p = .0009, p 2 = .063). In short, negative messages were most endorsed (via higher rankings) by negative feedback participants, and practically ignor ed by all ambiguous and positive feedback participants (regardless of response option condition), and nega tive feedback participants ranked negative messages higher than both positive and ambiguous feedback participants in both response option conditions (ps < .0001). Negative feedback pa rticipants were most likely to endorse negative messages when they were in the neutral message condition ( M negative feedbackMixed Messages = 10.78, M negative feedback Neutral Messages = 9.16, p <.0001)possibly endorsing negative messages here because they could not utilize th e negative interpretation of mixed messages to send aggressive messages. Ambiguous and positive feedback participants did not differentially endorse negative messages acros s response option conditions ( ps .84). This pattern of findings is consistent with the hypotheses that negative feedback participants would most strongly endorse negative messages relative to other feedb ack conditions, and also that negative feedback participants would curtail their use of negative messages when mixed messages were available. Regarding the ranking of positive statements, a feedback main effect ( F (2, 214) = 126.93, p < .0001), p 2 = .543), qualified by a Feedb ack X Option interaction ( F (2, 214) = 6.21, p =

PAGE 56

56 .0024, p 2 = .055) indicates that positive feedback participants and ambiguous feedback participants endorsed positive messages considerably more than negative feedback participants, regardless of response option (p s <.0001), but these positive f eedback and ambiguous feedback participants endorsed positive messages equally ( p Mixed Messages = .27, and p Neutral Messages = .24) However, while positive feedback particip ants were unaffected by response option ( p = .94), negative feedback participants in the mixe d messages condition endorsed positive messages more strongly than participants in the neutral messages condition ( p = .01). In contrast, the availability of mixed messages made ambiguous f eedback participants weaken their endorsement of positive messages compared to ambiguous feedback participants in the neutral messages condition ( p = .04). In sum, as predicted, positive feedback participants did employ primarily positive messages and were unaffected by the ava ilability of mixed messages, but ambiguous feedback participants employed positive messages just as much as positive feedback participants, which was not anticipated. However, ambiguous fee dback participants did behave consistently with the hypothesized no tion that they would endorse positive messages more strongly in the absence of mixed messages as a viable communication alternative, i.e., in the neutral messages condition. In other words, when ambiguous feedback participants had neutra l statements as their alternative response option, they endorsed outri ght positive messages more strongly than when mixed messages were available. In contra st, negative feedback participants gave weaker endorsements of positive messages when mixed me ssages were available than when neutral messages were available. In other words, negative feedback participants were nicer when mixed messages were a communication outlet than when they were not. Category Comparison To com pare the use of one category to another, three difference scores were calculated for the difference in the average ranks assigned to each category (i.e ., difference scores to compare

PAGE 57

57 alternative messages to negative messages, altern ative messages to positive messages, negative messages to positive messages). This way, relative endorsement of categories can be compared to see if participants endorse, for example, ne gative messages more or less than alternative messages. These difference scores were then submitted to the 3 (Feedback) X 2 (Response Option) general linear model described above. Table 3-2 presents average ranks and when differences between categories we re significantly different from one another. For expository purposes, Table 3-3 presents differe nce scores discussed below. It was predicted that positive messages would be endorsed more than negative or alternative messages for positiv e feedback participants, regardless of response option. In contrast, when mixed messages were available, it was expected th at both negative feedback and ambiguous feedback participants would endor se mixed messages most, with secondary endorsements of negative statements for negati ve feedback participants and positive messages for ambiguous feedback participants. In contrast, when neutral messages were the alternate form of communication instead of mixed messages, it was hypothesized that negative feedback participants would primarily endorse negative messages and am biguous feedback participants would primarily endorse positive and neutral statements. In general, participants in the positive feedback and ambiguous feedback conditions strongly endorsed positive messages, avoide d negative messages, and ranked alternative messages intermediately. This was true regard less of response option. However, negative feedback participants endorseme nt varied as a function of response option; mixed message participants rated positive messa ges and mixed messages equally highly and as more descriptive than negative messages. However, negative feed back participants in the neutral messages

PAGE 58

58 condition were more openly aggressive by rating negative messages more highly than positive messages (and as highly as neutral messages). More specifically, significant ma in effects of feedback were found for all three difference score comparisons (F negative to alternative (2, 214) = 25.66, p < .0001, p 2 = .193; F positive to alternative (2, 214) = 75.93, p < .0001, p 2 = .415; F negative to positive (2, 214) = 150.71, p < .0001, p 2 = .585). A significant main effect of option was also found for comparison of negative messages to alternative messages ( F (1, 214) = 8.08, p = .0049, p 2 = .036). These effects were qualified by significant Feedback X Option interactions for e ach of the three category comparison difference scores, ( F negative to alternative (2, 214) = 3.24, p = .04, p 2 =.029; F positive to alternative (2, 214) = 3.93, p = .02, p 2 = .035; F negative to positive (2, 214) = 7.68, p = .0006, p 2 = .070)2. In other words, differences between endorsing negative messa ges vs. alternative messages, positive vs. alternative messages, and positive vs. negative messages differed depending on feedback and response option for each kind of category compar ison. The significant orthogonal simple effects comparisons are presented in Table 3-2. Within each type of feedback and respons e option, participants consistently ranked negative messages as significantly less descriptive than alternative messages, with the sole exception that neutral message cond ition negative feedback participan ts did not differ in their use of negative messages and neutral messages ( p = .19). Participants in the positive feedback and ambiguous feedback conditions (in both res ponse conditions) ranked positive messages as significantly more descriptive than either negative messages or alternative messages ( ps < .0001). However, negative feedback participants in the mixed message condition did not differ in 2Although the three-way Category X Feedback X Option interaction was significant for ranking scores, the comparison of alternative stat ements to negative statements selection was only marginally significant, F (2, 216) = 2.63, p = .0741, p 2 = .024)

PAGE 59

59 their rankings of positive messages and mixed messages ( p = .12); negative messages were ranked as significantly less descriptive than both of these, p < .0001 and p = .01, respectively, indicating that while these negati ve feedback participants endo rsed negative statements more than positive or ambiguous feedback participants these participants still exhibited a slight preference for mixed messages and positive messages. Meanwhile, negative feedback participants in the neutral message condition ra nked positive messages as less descriptive than both neutral messages and negative messages ( p < .0001 and p = .0045, respectively) indicating that these participants alone were th e most nasty in their messages. This pattern of findings is part ially consistent with the pred icted effects. Specifically, as predicted, positive feedback participants did endorse positive messages more than negative or alternate messages, regardless of response option condition. The participants were motivated to be nice and overwhelmingly were. In contrast, ambiguous feedback pa rticipants were expected to use mixed messages more than either positive or negative messages in the mixed messages condition, but they employed positive messages mo re than mixed messages or negative messages in this condition. On the other hand, ambiguous f eedback participants use of primarily positive messages in the neutral messages condition was largely consistent with the hypothesis that participants would favor positive messages when mixed messages were not available. It was expected that negative feedback participants in the mixed me ssages condition would capitalize on the availability of mixed messages and e ndorse these more than negative or positive messages. While negative feedback participants did e ndorse mixed messages more than negative messages, they were not endorsed significantly more than positive messages in the mixed messages condition, indicating that perhaps the nega tive feedback participan ts were torn between positive messages and mixed messages. In contrast, it was hypothesized that because negative

PAGE 60

60 feedback participants in the ne utral messages condition could not satisfy their aggressive urges via mixed messages, they would endorse negative statements more strongly than the negative feedback participants with mixed messages available. Summary of Message Use The sim ple effects comparisons are presented in Table 3-2 for all of the above analyses. Orthogonal simple effects comparisons are presented in the table, with means having at least one factor in common and sharing a subscript not significantly different from one another. It was expected, generally, that participants would respond in kind to the feedback that they received. In other words, positive feedback participants would respond primarily in a positive manner, negative feedback participan ts would respond in a negative manner, and ambiguous feedback participants would respond vi a mixed messages when available, and would eschew outright negativity for positive messages when mixed messages were not available and instead, only neutral messages we re. Likewise, the ability to employ mixed messages should similarly curtail the use of negative feedback participants endorsement of outright negative messages. In general, participants receiving negative f eedback were considerably more aggressive overtly and covertly, relative to participants receiving ambiguous and positive feedback, as determined by the ranks assigned to negativ e messages and mixed messages. Ambiguous feedback participants did not respond as aggressively as wa s expectedparticipants responded quite similarly and in a generally positive fashion after receiving positive and ambiguous feedback. However, while positive feedback particip ants exhibited low levels of overt and covert aggression and were entirely unaffected by res ponse option, ambiguous feedback participants were somewhat affected by response option (mixed messages versus neutral messages as the alternative message). As compar ed to their neutral message counterparts, ambiguous feedback

PAGE 61

61 participants were less nice when they had mi xed messages available to them, but negative feedback participants were nicer when they had mixed messages available to them. Thus, while Hypothesis IX.1 predicted that ambiguous fee dback participants w ould use more mixed messages than positive feedback or negative fee dback participants, this was not supported as negative feedback participants were mo st likely to endorse mixed messages. Because response option (mixed messages ve rsus neutral message conditions) had no effect on rankings of negative messages or alternate messages made by positive feedback participants ( ps > .90), Hypothesis VII.2 (that positive feedback participants would not differ in their use of positive and negative messages ac ross response options) was supported. Positive feedback participants were always nice and unaggre ssive, regardless of whether they were in the mixed message or neutral message condition. Because negative feedback participants were generally nicer in the mixed messages condition than the neutral messages condition, but the reverse was true for ambiguous feedback participants, Hypothesis II (that ov erall aggression would be less wh en one is forced to aggress openly than when the opportunity to aggress covert ly is also available) received mixed support. Negative feedback participants did not follo w Hypothesis II: these participants were less negative when they had mixed messages available to them and did not differ in their usage of alternate statements. Ambiguous feedback participants, however, ranked mixed messages more highly than neutral messages (attemp ting to be covertly aggressive ) while they did not differ in their use of negative messages. Hypothesis IX.3 predicted that ambiguous feedback participants in the mixed messages condition would use many mixed messages and few, if any, negative messages. In comparing the use of negative messages to the alternate messages, participants in each of the feedback and

PAGE 62

62 response option conditions ranked negative message s as less descriptive than the alternate messages, except for negative feedback pa rticipants in the neutra l messages condition, who did not differ in their rankings of negative messages and neut ral messages, thus supporting Hypothesis IX.3. Likewise, Hypothesis IX.4 predicte d that ambiguous feedback participants in the neutral messages condition would send primaril y neutral and positive messages, and received qualified support. While positive and ambiguous f eedback participants ranked mixed messages and negative messages similarly, negative feedba ck participants ranked negative messages and mixed messages more similarly than both the am biguous feedback participants, and the positive feedback participants. This effect stems primarily from negative feedb ack participants ranking mixed messages higher than positive feedback or ambiguous feedback participants. In other words, while ambiguous feedback and positive feedback participants demonstrated a clear preference for mixed messages over negative message s, negative feedback participants showed the same preference to a much lesser degree, s eeing a considerably sma ller distinction between mixed messages and negative messages. One interpretation of the negativ e feedback participants behavi or is that when negative feedback participants did not have the opportunity to aggress covertly, they shifted to overt aggression. In contrast, negative feedback participants who could aggress covertly did so and reduced their level of overt aggression. This primary endorsement of positive messages by positive feedback participants (rather than negative messages or mixed messages) supports Hypothesis VII.1 that participan ts receiving positive feedback w ould primarily endorse positive messages. In contrast, Hypothesis VIII was not fully supported. Specifically, Hypothesis IX.4 predicted that negative feedback participants in the neutral messages condition would send primarily neutral messages and few negative messag es, and this was found not to be the case.

PAGE 63

63 Instead, while the general trend was that neutra l messages were endorsed more strongly than negative messages by those in the neutral messages condition, this difference was not statistically significant. However, Hypothesi s VIII.2 predicted that negati ve feedback participants with mixed messages available to them would primarily endorse mixed messages over than negative messages, and this was indeed found to be the case. In sum, ambiguous feedback and positive feedba ck participants responded similarly and in a generally positive manner to their feedback such that they primarily endorsed positive messages, followed by alternate messages, and most weakly endorsed negative messages, which were essentially ignored. The only real diffe rence between positive and ambiguous feedback participants lies in the manner in which they were affected by the availabi lity of mixed messages. Specifically, positive feedback participants were not at all affected by their response option condition, but ambiguous feedback participants endorsed mixed messages more than neutral messages, possibly to send aggressive overtones, but also possibly to send positive intimations. In other words, the availability of mixed messages seems to have made ambiguous feedback participants slightly less nice, as further in dicated by their weakened endorsement of positive messages when mixed messages were available. In contrast, for negative feedback participants, the availability of mixed messa ges resulted in less negative message endorsement and stronger positive message endorsement than the availability of neutral messages. Objective Assessment of Design and Evaluation Quality It is reasonable to ask whether som e of the effects might be due to variations in how well participants completed their own design or how positively participants had originally evaluated their partners. For example, pa rticipants who had evaluated their partner very positively and received negative feedback or ambiguous feedback may have reacted more strongly to feedback than participants who had not ev aluated their partner so positively. Likewise, participants who

PAGE 64

64 had developed a particularly exceptional squirre l-proof birdfeeder may have responded more strongly to negative or ambiguous feedback than participants who did not create such an exceptional design. Two independent raters evaluated each participants squirr el proof birdfeeder design on four different dimensions: 1) how much effort th e participant exhibited in creating the design, 2) how creative the design was, 3) how effectiv e the design would be at keeping squirrels out of a birdfeeder, and 4) how practical the design wa s (intraclass correlation reliabilities, ICC (3,k): r = .69, .66, .58, and .29, respectively). The two raters we re assumed to be the entire population of raters and the reliability is for the mean of th eir ratings (Shrout & Flei ss, 1979). The raters also evaluated the written evaluation sent to the nonexistent partner of the partners own creative intelligence assessment on two dimensions: 1) how positive the evaluation was, and 2) how much effort the participant put in to evaluating the partners work ( = .78 and .78, respectively). None of these measures significan tly moderated the above findings. Interaction Expectations The m anipulation of interaction expectations was successful. Participants who were told they would be meeting their partner should have reported that they were going to meet someone more than participants who were told they would not be meeting their partner. A betweensubjects t-test revealed that this was the case, t (142) = 8.70, p < .0001, M meet = 1.99, M NO meet = 1.47. Anticipating a future interaction was genera lly expected to rest rain the use outright negative messages in favor of mixed messages such that participants would shift from mixed messages to negative messages when not facing a meeting such that outright negative messages would be used less and mixed messages would be used more when a meeting was imminent than when no meeting was anticipate d. To assess the actual impact of expecting an important,

PAGE 65

65 imminent collaborative interaction with their pa rtner on the likelihood of using various forms of communication (positive, covert aggression, straightforward aggression), the participants in the offset comparison conditions were included in a 2 X 2 Gene ral Linear Model: Feedback (Ambiguous vs. Negative) X Interaction Expect ation (Meet vs. No Meet), looking at each statement category ranking (positive, negative, mixed messages). All participants in this analysis had mixed messages available to them. A marginal Feedback X Expectation interact ion was found for the ranking of the positive messages ( F (1, 137) = 2.72, p = .10). As shown in Table 3-4, negative feedback participants more strongly endorsed positive messages if they anticipated a future interaction ( F (1, 137) = 4.06, p = .0460, M meet = 9.31, M NO meet = 10.30), whereas ambiguous feedback participants did not differ in their ranking of positive messages as a function of anticipated interaction ( F (1, 137) = 0.04, p = .85). See Table 3-4 for mean rankings as a function of feedback and meeting expectation. Rankings of mixed messages and negative messages were unaffected by interaction expectation ( p ambiguous feedback = .30, p negative feedback = .44), or an interaction between interaction expectation and feedback ( p ambiguous feedback = .16, p negative feedback = .47). The general lack of impact of an expected meeting fails to s upport Hypothesis I. In addition, because negative feedback participants not expe cting a future interaction primarily endorsed mixed messages rather than negative messages, Hypothesis VIII. 3 was not supported, which predicted that few mixed messages would be endorsed, and particip ants would instead rely on negative messages. Finally, Hypothesis IX.2 predicted that ambiguous feedback participants not expecting an interaction would use more mixed messages than those expecting an interaction and would primarily endorse negative messages. However, because ambiguous feedback participants expecting an interaction did not differ in their use of mixed messages compared to ambiguous

PAGE 66

66 feedback participants not expecting an interaction and sent primarily positive messages rather than negative messages, Hypothesis IX.2 was not supported. Perceptions of Partner Twenty-three questions from the Pre liminary Interaction Assessment assessed participants perceptions of thei r partner and reactions to the test, and these questions were grouped into nine factors including how likeable, smart, condescending, straightforward, sincere, and a skilled communicator the other person was as well as how liked they expected to be by their partner, how smart they considered themselves to be, and how valid they believed the test to be. See Table 3-5 for a list of the items in each trait/grouping and relevant alphas. It was generally predicted that ambiguous feedback participants not expe cting an inter action would use more mixed messages than those expecting an in teraction and would prim arily endorse negative messages. Main effects of feedback were found for how likeable straightforward, sincere, skillfully communicative, smart and condescending ones partne r was such that, in ge neral, participants receiving positive feedback perceived their partner to be more likeable, sincere, a skilled communicator, smarter and less condescending th an partners sending negative feedback. See Table 3-6 for F and p-values, effect sizes, mean s and simple effects comparisons. Significant main effects of feedback were also found on how liked participants felt by their partners, how smart participants themselves felt and how valid particip ants believed the test to be. Participants receiving positive feedback thought they were li ked more by their partners, thought the test was more valid, and thought they were themselves smarter than particip ants receiving negative feedback. Negative feedback par ticipants believed their partner was more straightforward than participants receiving positive feedback ( p = .0015) or ambiguous feedback ( p = .0008). Ambiguous feedback partners were not seen as significantly different from positive feedback

PAGE 67

67 partners on any of these measures. In short, partners who delivered positive feedback and partners who delivered ambiguous feedback were seen positively and created positive feelings, and both differed from partners who delivered negative feedback. Hypothesis X.1 and X.2 predicte d that positive feedback part ners would be most likeable and would seem to like the participants themselves most while little difference would be found between negative and ambiguous feedback partne rs. While positive feedback partners were viewed more favorably than negative feedback partners, ambiguous feedback partners were viewed just as favorably, thus failing to s upport the notion that ambi guous feedback partners would suffer a tarnished reputation for their use of covert aggression. Hypothesis XI.1, XI.2, and XI.3 predicted that ambiguous feedback partners would be seen as the best communicators, followed by positive feedback partners, and negative feedback partners would be seen as the least effective communicators. Be cause positive and ambiguous fee dback partners were viewed as equally good communicators a nd significantly better communicators than negative feedback partners, Hypothesis XI.1 and XI.3 were both supported (that ambiguous feedback partners would be seen as more skillful communicators than negative feedback partners and positive feedback communicators may experien ce a halo effect and be seen as a better communicator than negative feedback communicators). However, because ambiguous and positive feedback partners were viewed similarly, Hypothesis XI.2 (that ambiguous feedback partners would also be seen as more skillful communicators than positive f eedback partners) was not supported. Therefore, although technically Hypothesis XI received partial support in th at both positive and ambiguous feedback partners would be seen as be tter communicators than negative feedback communicators, the main focus of this hypot hesis predicted that ambiguous feedback communicators would be seen as particularly cu nning communicators in their use of covert

PAGE 68

68 aggression, even relative to positive feedback communicators, and this was not found to be the case. Hypothesis XII predicted that participants receiving (1) negative and (2) ambiguous feedback would perceive the creative intelligence test as less valid than positive feedback participants. In actuality, ambiguous and positive feedback participants perceived the creative intelligence test to be equally valid, and significantly more valid than negative feedback participants, thus Hypothesis XII.1 was suppor ted but not Hypothesis XII.2. Hypothesis XIII predicted that negative and am biguous feedback participants would not internalize the disparaging feedback they received and should not differ in how creatively intelligent they believe themselves to be (although negative feedb ack participants would be more likely to internalize than ambiguous fee dback participants). Although ne gative feedback participants reported that they believed themselves to be less creatively intelligent than ambiguous feedback participants and positive feedback participants this was not drastical ly different, and the ambiguous feedback participants did not differ fr om the positive feedback participants in how creatively intelligent they believed themselves to be. Thus, the spirit of this hypothesis was generally supported. An additional measure was included to assess whether participants varied regarding how positively they believed they had evaluated their partners creative intelligence assessment prior to completing the First Impressions Questionnaire. Participants did not differ in how positively they believed they had evaluated their pa rtners work as function of feedback ( F (2, 219) = 0.04, p = .96), response option ( F (1, 219) = 0.99, p = .32), or a Feedback X Option interaction (F (2, 219) = 0.82, p = .44).

PAGE 69

69 To assess whether perceptions of their partne rs were affected by whether participants expected an interaction, a 2 (In teraction Expectation: Meeting or No Meeting) X 2 (Feedback: Ambiguous or Negative) general linear model wa s conducted with the offset comparison groups to assess whether meeting expectations predic ted or moderated the above findings. With two exceptions, interaction expectations did not predict or moderate how participants viewed their partners. Participants who expected to meet their partner perceive d the test to be somewhat less valid than those who did not e xpect to meet their partner ( F (1, 139) = 6.12, p = .01, p 2 = .042, M meet = 5.15, M NO meet = 5.76). In addition, there was a marginally significant trend for participants who expected a m eeting to report that they had evaluated their partner more positively ( F (1, 139) = 3.24, p = .07, p 2 = .023, M meet = 7.49, M NO meet = 7.09) suggesting that perhaps participants who anticipated an interactio n with someone wished to be a bit nicer than those who did not expect to meet their partner. Affect To determ ine whether or not affect differe d as a function of feedback, positive and negative affect scores (Cronbachs = .87 and .86, respectively) were subjected to a model with feedback and response option as predictors. Positiv e and negative affect scor es were slightly, but significantly correlated, r (225) = -.18, p = .005. Mean levels of affect by feedback are presented in Table 3-6 along with F and p-values and effect sizes. A main effect of feedback was found for both positive and negative affect. Negative feedback participants reported less positive affect and more negative affect than positive feedback or ambiguous feedback participants ( p < .0001); the latter two conditions did not di ffer. Therefore, Hypothesis XIV .1 and XIV.2 were supported (that positive feedback participants would report positive emotions, and negative feedback participants would report nega tive emotions) but Hypothesis XIV.3 (ambiguous feedback participants would report both positive and negative emotions) was not supported

PAGE 70

70 To assess whether participants experienced diffe rential affect as a result of expecting an interaction, a 2 (Interaction Exp ectation: Meeting or No Meeti ng) X 2 (Feedback: Ambiguous or Negative) general linear model was conducted w ith the offset comparison groups to assess whether meeting expectations predicted or m oderated the general a bove findings regarding positive and negative affect. Interaction expectat ion did not predict either positive or negative affect, alone or in combination with feedback. Communication Preference Participants were asked to select a met hod of communicating for their collaborative interaction with their partner. To test the hypothesis that participants r eceiving negative or am biguous feedback would eschew face-to -face communication, communication method preferences were submitted to a chi-square analys is with feedback. As shown in Table 3-7, the results indicate a significant variatio n in communication method by feedback ( 2 = 19.85, p = .0005), such that face-to-face communication was preferred by all feedback conditions, but negative feedback participan ts preferred computer-mediated communication more than ambiguous feedback participants or positive f eedback participants. Therefore, Hypothesis XV (that participants receiving negativ e and ambiguous feedback would be most likely to avoid faceto-face interactions and instead seek out le ss immediate forms of communication such as computer mediated interaction and intercom mediated interaction), found support in negative feedback participants, but not among ambiguous f eedback or positive feedback participants. Moderators A variety of individual difference variables were included as possible m oderators of effects. To determine the nature of any inte ractions, analyses were conducted using the SAS General Linear Model (GLM) proc edure on ranking scores. The pr edictors in the model were feedback (3 levels: ambiguous, negative, or positive), option (2 levels: mixed message or neutral

PAGE 71

71 message condition), each respective moderator (as a continuous, centered variable), and all interactions. Moderators were cen tered so that any main effects of option or feedback could be interpreted as an effect on a person who has an average score on the moderator. High and low levels of each moderator were defined as 1 SD above and below the mean level; scores reported for high, low, and average moderator reflect GL M estimates of the values of the dependent measures at these integrity levels (see Cohen, Cohen, West & Aiken, 2003). Perspective Taking It was expected that participants higher in perspective taking woul d show an increased likelihood of using covert aggression, i.e., endor sing mixed messages, due to previous research findings that perspective takers were more capab le of indirect aggressi on than low perspective takers. Neither total empathy scores ( = .78) nor empathic concern scores ( = .76) produced any significant affects on ranking of statements, al one or in combination w ith other predictors. A three-way feedback X option X perspective taking interaction was found on ranking negative messages, F (2, 208) = 4.81, p = .0091, p 2 = .044, perspective taking = .76. See Figure 3-1 for estimated mean rankings generated by the SAS estimate command for participants high, medium, and low in perspective taking as a func tion of feedback and option. Table 3-8 presents these means along with comparisons across condi tions and levels of perspective taking. Closer examination of these interactions reveals that participants receiving positive or ambiguous feedback were unaffected by either re sponse option or perspective taking ability in their selection or ranking of negative messages. In contrast, negative feedback participants generally ranked negative messages higher in the neutral messages condition than in the mixed messages condition and this tendency was exagge rated among participants high in perspective taking. Hypothesis V predicted that participants higher in perspective taking would use more mixed messages. This was not the case as perspe ctive taking did not pred ict rankings of mixed

PAGE 72

72 messages. However, perspective taking did predic t rankings of negative messages in response to negative feedback such that, in the mixed messa ges condition, high perspe ctive takers did not endorse negative messages as strongly as low perspective takers. On the other hand, in the neutral messages condition, high perspective ta kers endorsed negative messages much more strongly than low perspective takers. Perhaps hi gh perspective takers in the mixed messages condition perceived their endorse ment of any mixed messages as sending negative messages and so could back off the endorsement of negative messages. In contrast, because neutral messages did not send this aggressive undertone, participants who wished to aggress were forced to more strongly endorse negative messages in the neutral message condition. Other Moderators Several other possible moderators were tested including hostility ( = .78), anger ( = .82), physical aggression ( = .85), direct aggression ( = .87), and indirect aggression ( = .83). These m oderators produced no significant effects either alone or in in teraction with other predictors. As such, the results failed to suppor t Hypothesis IV (that par ticipants with greater proclivity toward direct aggres sion would show higher rates of ove rt aggression), and failed to support Hypothesis III.1, III.2, and III.4 (that participants with gr eater proclivity toward hostility, anger, and indirect aggression would show higher rates of covert aggression). Verbal aggression ( = .69), was found to significantly pred ict ranking of alternate messages ( F (1, 208) = 4.61, p = .03) such that less verbally aggressive participants endorsed alternate messages more than more verbally aggressive participants3, thus failing to support Hypothesi s III.3 that more verbally aggressive participants would s how higher rates of covert aggr ession. Proclivity toward verbal aggression did not predict rank ings of negative messages. 3 Selection of alternative statements (a s opposed to ranking) was only marginally predicted by proclivity to ward verbal aggression, F (1, 210) = 1.94, p = .17.

PAGE 73

73 It was expected that individuals high in social desirability would be less likely to endorse negative or mixed messages than positive me ssages, When these high social desirability participants do act on their aggressive motives, though, they would endorse mixed messages more than low social desirability participants, who would prefer outright negative messages over mixed messages Social desirabi lity significantly interacted with feedback and option in ranking of negative messages (F (2, 208) = 4.33, p = .01, p 2= .040), and ranking of positive messages ( F (2, 208) = 3.01, p = .0513, p 2 = .028).4 See Figures 3-2 and 3-3 for mean rankings of negative and positive messages. Table 3-9 presents these means along with comparisons across conditions and levels of perspective taki ng. The interaction indicates th at while negative feedback participants rankings were affected by soci al desirability and response option, positive and ambiguous feedback participants generally were not. Specifically, negative feedback participants in the mixed messages condition endorsed nega tive messages less than negative feedback participants in the neutral messages condition. This general tendenc y was exaggerated among high social desirability participants in the mixed messages condition such that high social desirability participants actually endorsed nega tive messages significantly more than low social desirability participants. The inverse numerical pattern is true for rankings of positive messages, consistent with the above pattern that high social desirability partic ipants are most likely to engage in socially desirable behavior when mixed messages were available rather than neutral messages as the alternate message. In keeping w ith the general findings, negative feedback participants endorsed 4 The effects of social desirability and pe rspective taking reflect similar patterns on the dependent measures. Social desirability was si gnificantly correlated wi th perspective taking, r (227) = 0.33. However, this correlation alone cannot account for the similarities as a similar strength correlation was found between other moderators, e.g., social desirability was significantly correlated with both verbal aggression and hostility ( r (227) = -.31 and .45, respectively, and these moderators did not si gnificantly predict the dependent measures.

PAGE 74

74 positive messages more in the mixed messages condition than the neutral messages condition, but this was particularly true for those high in social desirability. In sum, participants higher in social desirability pursued thei r socially desirable motives to be nice in the face of negative feedback when mixed messages were available, but when only neutral messages were available, high social desirability participants were significantly more negative than their low social desirability c ounterparts. Therefore, Hypothesis VI.1 and VI.2 were not supportedsocial desira bility participants did not di ffer in their use of covert aggression compared to low socially desirability participants. However, the presence or absence of mixed messages as an option did affect pa rticipants ranking of negative and positive messages. In the wake of having mixed messages av ailable, high social desirability participants were considerably nicer than their low social desira bility counterparts. In c ontrast, in the wake of having neutral messages available, high social desi rability participants were considerably more aggressive than low social desirability participants.

PAGE 75

75 Table 3-1. Pilot study: Perceptions of feedback Ambiguous feedback Negative feedback Positive feedback Positive Interpretation 7.00a (1.70) 1.92 (1.26) 7.68a (1.25) Negative interpretation 3.46 (1.53) 7.42 (1.45) 2.12 (0.88) Double-meaning 5.35 (2.37) 2.85b (2.34) 3.68b (2.06) Would be upset 3.27 (2.27) 5.15 (2.29) 2.04 (2.07) Other trying to be superior 5.76g (2.22) 6.38g (2.23) 3.78 (2.15) Other is concealing insult 4.92e (1.98) 4.23ef (2.41) 3.13f (1.66) Other is sincere 5.38e (2.23) 5.96cd (2.07) 6.44d (1.53) n 26 26 25 For each dependent measure, means sharing a subscript are not significantly different.

PAGE 76

76 Table 3-2. Ranking means Feedback Statement type Mixed messages Neutral messages Ambiguous feedback Alternative messages 9.18a (1.07) 10.01b (1.43) Negative messages 13.40cd (0.48) 13.47ce (0.15) Positive messages 5.92f (1.22) 5.01g (1.46) n 45 36 Negative feedback Alternative messages 8.41hi (1.72) 8.67hj (1.55) Negative messages 10.78 (2.40) 9.16j (2.90) Positive messages 9.31i (3.05) 10.66 (2.44) n 29 35 Positive feedback Alternative messages 9.66ak (1.40) 9.64bk (1.62) Negative messages 13.39dl (0.38) 13.44el (0.27) Positive messages 5.45fm (1.43) 5.42gm (1.67) n 38 37 Values reflect average ranking of messages. Therefore, lower values reflect more endorsement of that category. Orthogonal simple effects comparis ons are presented in the table, with means having at least one factor in common and sharing a subscript not significantly different from one another. Standard deviations are indicated in parentheses.

PAGE 77

77 Table 3-3. Category ranking difference scores Feedback Statement type Mixed messages Neutral messages Ambiguous feedback Alternate minus Negative 4.22 (1.12) 3.46 (1.42) Negative minus Positive -7.48 (1.51) -8.46 (1.50) Positive minus Alternate 3.26 (2.24) 5.00 (2.89) Negative feedback Alternate minus Negative 2.37 (2.85) 0.49 (3.95) Negative minus Positive -1.47 (5.22) 1.50 (5.13) Positive minus Alternate -0.90 (4.33) -1.99 (2.89) Positive feedback Alternate minus Negative 3.73 (1.47) 3.79 (1.61) Negative minus Positive -7.94 (1.56) 8.02 (1.77) Positive minus Alternate 4.21 (2.81) 4.23 (3.28) Values represent differences in category ranki ngs. In the first cell, 4.22 indicates that participants receiving ambiguous feedback in the mixed messages condition rated mixed messages 4.22 ranks above negative messages, on aver age. Standard deviations are indicated in parentheses.

PAGE 78

78 Table 3-4. Ranking means with offset comparis on groups not expecting a future interaction Feedback Statement type Future meeting No future meeting Ambiguous feedback Mixed messages 9.18a (1.07) 9.28a (1.21) Negative messages 13.40b (0.48) 13.39b (0.32) Positive messages 5.92c (1.22) 5.84c (1.11) n 45 36 Negative feedback Mixed messages 8.41de (1.72) 7.80d (1.97) Negative messages 10.78f (2.40) 10.40fg (2.06) Positive messages 9.31e (3.05) 10.30g (2.08) n 29 31 Values reflect average ranking of messages. Therefore, lower values reflect more endorsement of that category. Orthogonal simple effects comparis ons are presented in the table, with means having at least one factor in common and sharing a subscript not significantly different from one another. Standard deviations are indicated in parentheses.

PAGE 79

79Table 3-5. Trait groupings and relevant Cronbach alphas Trait Liking of the other .91 Do you believe that you will like the other participant? Do you think that the other participant is generally friendly? Do you believe you will respec t the other participant? How much do you think you will enjoy work ing with the other participant? How well do you think this person gets along with other people? How much are you worried about inte racting with the other person?* Intelligence of the other .82 How much creative intelligence do you th ink the other participant possesses? How much do you value the other part icipants opinion regard ing creative intelligence problem solving? Evaluation by the other .95 How much creative intell igence does the other par ticipant believe you have? How positively did the other person evaluate your work? Straightforwardness of the other To what extent do you think that the other person is straightforward? Sincerity of the other To what extent do you think that the other person is sincere? Communication skill of the other To what extent do you think that the ot her person is a skilled communicator? Liking by the other .91 Do you believe that the othe r participant will like you? How much do you think the othe r participant respects you? Test validity .88 How well do you think the first assessment of creative in telligence actually assesse d creative intelligence? How well do you think the second assessment of creativ e intelligence will assess you r creative intelligence? Self intelligence How much creative intelligence do you believe you possess? Evaluation of others individual cr eative intelligence assessment How positively did you evaluate the other persons work? *indicates item was reverse scored

PAGE 80

80Table 3-6. Perceptions of partner and affect Dependent measure Ambiguous feedback Negative feedback Positive feedback Error df F p p 2 Liking of the other 6.96a (0.87) 4.02 (1.18) 7.16a (0.58) 221 265.85 <.0001.706 Intelligence of the other 7.33a (1.13) 4.28 (1.62) 7.15a (0.89) 219 131.44 <.0001.546 Evaluation by the other 2.58a (1.21) 8.42 (0.72) 2.34a (0.68) 219 943.86 <.0001.896 Straightforwardness of the other 6.83a (1.43) 7.55 (1.25) 6.87a (1.02) 221 7.06 .0011 .060 Sincerity of the other 6.78a (1.26) 4.64 (2.11) 6.94a (1.05) 221 50.07 <.0001.312 Communication skill of the other 7.20a (1.33) 4.70 (1.95) 7.17a (0.95) 221 68.41 <.0001.382 Liking by the other 6.68a (0.98) 3.37 (1.27) 6.67a (0.99) 221 219.25 <.0001.662 Test validity 6.43a (1.47) 4.38 (1.71) 6.69a (1.04) 219 54.63 <.0001.333 Self intelligence 6.70a (1.15) 5.80 (1.68) 6.58a (1.32) 219 8.79 .0002 .074 Evaluation of others assessment 7.36a (1.17) 7.32a(1.41) 7.38a (1.10) 219 0.04 .9566 .000 Positive affect 27.94a (6.89) 23.03 (7.75) 29.21a (7.32) 220 13.44 <.0001.109 Negative affect 13.46a (5.36) 18.23 (6.41) 12.58a (3.07) 220 24.48 <.0001.182 For each trait or affect, means that share a subs cript are not significantly different. Subscripts a and c denote marginally significant differences, p = .15 and p = .11, respectively. Standard deviati ons are indicated in parentheses.

PAGE 81

81 Table 3-7. Participant communicatio n method preferences Chosen communication method Feedback Face-to-face Speaker system Computer Total Ambiguous 80 0 3 83 (96.39%) (0.00%) (3.61%) Negative 47 2 15 64 (73.44%) (3.13%) (23.44%) Positive 67 1 5 73 (91.78%) (1.37%) (6.85%) Values not in parentheses indicate number of individuals choosing each communication method. Values in parentheses indicate percentage of pa rticipants in each fee dback condition selecting that particular communication method.

PAGE 82

82Table 3-8. Ranking of negative messages as a function of pe rspective taking, feedback, and response option Feedback Mixed messages Neutral messages Response option comparison Ambiguous feedback Perspective slope: ns ns 13.36 13.49 ns 13.42 13.47 ns High Average Low 13.47 13.46 ns Negative feedback Perspective slope: ** 11.34 8.01 **** 10.79 8.82 **** High Average Low 10.25 9.63 ns Positive feedback Perspective slope: ns ns 13.45 13.46 ns 13.36 13.44 ns High Average Low 13.28 13.42 ns For each level of feedback and perspective taking, the response option comparison is indicated in the far right column ****indi cates p <.0001, ** indicates p < .01, ns indicates p > .05. The significance of perspective taking for each combination of response option and feedback is indicated each set of means. Negative feedback di ffered significantly from both posi tive and ambiguous feedback for both response options at all levels of perspective taking; positive feedback did not differ significantly from ambiguous feedback at any level of perspective taking for either response option.

PAGE 83

83 Table 3-9. Ranking of positive and negative messages as a function of social desirability, feedback, and response option Feedback Mixed messages Neutral messages Response option comparison Ambiguous feedback Negative ranking Social des. slope ns ns High 13.4813.48 ns Average 13.4013.47 ns Low 13.3213.46 ns Positive ranking Social des. slope ns ns High 5.835.12 ns Average 5.924.98 Low 6.014.84 .06 Negative feedback Negative ranking Social des. slope ** High 11.848.56 **** Average 10.959.13 **** Low 10.069.70 ns Positive ranking Social des. slope ns High 8.2710.98 *** Average 9.1410.68 *** Low 10.0110.38 ns Positive feedback Negative ranking Social des. slope ns ns High 13.4213.46 ns Average 13.4013.44 ns Low 13.3713.41 ns Positive ranking Social des. slope ns ns High 5.655.05 ns Average 5.475.45 ns Low 5.285.85 ns Within each level of feedback and social desirability, the response option comparison is indicated between the column s of means. ****indicates p <.0001, ***indicates p < .001, *indicates p <.05, ns indicates p > .05. The significance of social desirability for each combination of response option and feedback is indicated above each set of means. For all three dependent measures (negative ranking and pos itive ranking) negative feedback differed significantly from both positive and ambiguous feedback for both response options at all levels of social desirability; positive feedback did not differ significantly from ambiguous feedback at any level of social desirabili ty for either response option.

PAGE 84

84 Negative Message Ranking 7 8 9 10 11 12 13 14 Low Average High Perspective Taking Average Ranking Ambiguous Feedback Mixed Messages Ambiguous Feedback Neutral Messages Negative Feedback Mixed Messages Negative Feedback Neutral Messages Positive Feedback Mixed Messages Positive Feedback Neutral Messages Figure 3-1. Mean ranking for ne gative messages as a function of perspective ta king, feedback, and response option. Lower value s indicate more endorsement of negative messages.

PAGE 85

85 Negative Message Ranking 7 8 9 10 11 12 13 14 Low Average High Social Desirability Average Ranking Ambigous Feedback Mixed Messages Ambiguous Feedback Neutral Messages Negative Feedback Mixed Messages Negative Feedback Neutral Messages Positive Feedback Mixed Messages Positive Feedback Neutral Messages Figure 3-2. Mean ranking for ne gative messages as a function of social desirability, feedback, and response option. Lower valu es indicate more endorsement of negative messages.

PAGE 86

86 Positive Message Ranking 4 5 6 7 8 9 10 11 Low Average High Social Desirability Average Ranking Ambiguous Feedback Mixed Messages Ambiguous Feedback Neutral Messages Negative Feedback Mixed Message Negative Feedback Neutral Messages Positive Feedback Mixed Messages Positive Feedback Neutral Messages Figure 3-3. Mean ranking for positive messages as a function of social desirability, feedback, and response option. Lower valu es indicate more endorsement of positive messages.

PAGE 87

87 CHAPTER 4 DISCUSSION It was predicted that par ticipants receiving am biguous feedback conveying a mixed message would interpret this feedback as covert aggression, and these part icipants would prefer to respond to such a message with covert aggression themselves by endorsing mixed message statements. In contrast, it was pr edicted that recipients of ambi guous feedback who were not able to respond with their own mixed message statemen ts, would be cautious and resort to primarily positive messages. However, participants did not react to ambiguous feedback as expected and instead responded to their partners in a fashion quite similar to the manner in which positive feedback participants responded to their partnerin a primarily positive manner. The availability of a mixed message option did produce some subtle effects on ambiguous feedback and negative feedback recipients. Specifically, while ambiguous feedback participants did not endorse mixed messages very strongly, th ey did endorse mixed messages more strongly than neutral messages (with a compensating we akened endorsement of positive messages), indicating some sensitivity to the message option manipulation and the subtle differences between neutral and mixed messages. One possibl e interpretation of this pattern is that ambiguous feedback recipients were indeed sensiti ve to the duality of both the feedback they received and the messages they sent. If so, this pattern of message endorsement indicates that these participants did perceive the feedback as a possible slight and we re responding, subtly, via slightly stronger endorsement of mixed messages. Ho wever, this possibility is not consistent with other measures. Ambiguous feedback participants felt similar levels of positive and negative affect as positive feedback participants, indicating that they were not more upset by the partners assessment of their work than positive feedb ack participants. Moreover, ambiguous feedback

PAGE 88

88 participants reported perceiving their partner as positively as positive feedback participants saw their partner. Although it is possible that ambiguous feedback participants did perc eive the duality in their feedback and employed the ne gative interpretation of the mixe d messages in retaliation, this is unlikely. Instead, what is more consistent with the other reports from the ambiguous feedback participants is that they were as motivated as positive feedback participants to send positive messages, and endorsed mixed messages because of their positive interpretation while oblivious to the negative interpretation. In short, their a ttempt to be positive was diverted between positive messages and mixed messages. The same attempt to send positive messages would not cause a strong endorsement of neutral messages because neutral messages were clearly less positive than positive messages and the positive aspect of mixed messages. This interpretation is consistent with the finding that positive feedback and ambig uous feedback participants did not significantly differ from one another in their use of positive messages, negative messages, or mixed messages. The only real difference between positive feedba ck and ambiguous feedback participants was that ambiguous feedback participants were somewhat affected by the mixed message option whereas positive feedback participants were not. In contrast, while negative feedback participants endorsed mixed messages and neutral messages equally, there are indications that negati ve feedback participants may have still been sensitive to the dual nature of the mixed messages. More specifically, after negative feedback, participants were more negativ e and less positive in their messages if neutral message options were available than if mixed message options we re available. On the surface, then, those who had mixed message options seemed friendlier. Yet, regardless of the message option condition, those who received negative feedback experi enced negative affect, rated the other person

PAGE 89

89 negatively, and clearly dis liked that other. This suggests that the availability of the mixed messages may have led participants to become less hostile, which seems unlikely given their later negative affect and ratings of the other, or that they used the mixed messages as digs. The latter interpretation is consistent with the ratings of the pilot participants and the intent of the manipulation. If this latter interpretation is corr ect, then these participants may have then felt endorsing positive messages better disguised their true motivation for endorsing mixed messages. Moreover, after receiv ing negative feedback, these part icipants may have been more attune to the negative connotati ons of mixed messages whereas positive and ambiguous feedback participants were not as tuned in to the possible negative connot ations of the mixed messages. Thus, ambiguous and positive feedback participan ts may have been less aware of the negative connotations of the negative messages. Although negative and ambiguous feedback participants were subtly influenced by the message option manipulation, positive feedback part icipants were not at all affected by this manipulation. Participants receiving positive feedback did not differ in their use of positive messages, negative messages or alternate messa ges as a function of message option. Instead, positive feedback participants chose to be prim arily positive in their messages sent to their partner, avoided overtly nega tive messages altogether, and di d not treat mixed messages and neutral messages differently, ranki ng these two types of messages similarly to each other; these messages were ranked intermediate between negative messages and positive messages. Comparisons between Positive and Ambiguous Feedback It m ay seem, on the face of it, that particip ants did not perceive much difference between the positive feedback and ambiguou s feedback. However, pilot te sting clearly indicated that ambiguous feedback was seen as significantly more upsetting, more negative, as conveying a negative meaning at the same time as a positive meaning, as concealing more of an insult, being

PAGE 90

90 less sincere, and as more of an attempt to be superior. If anything, pilo t testing indicated that participants should have perc eived the ambiguous feedback as much more like the negative feedback than the positive feedback. There are at least two possible in terpretations of this pattern of relatively few differences between positive feedback and ambiguous feedback in the face of pilot ev idence that indicates there should be many differences between these two feedback conditions. One possibility is that although participants clearly identi fied the multiple nuanced interpre tations of the feedback, they preferred to play it safe. In other words, although the ambiguous, covertly aggressive feedback was upsetting, in deciding how to respond, part icipants considered the consequences of responding aggressively (covertly or overtly) an d being wrong as more dire than responding nicely and being wrong. In keeping with this line of reasoning, ambi guous feedback participants may have identified the possible dual interpretations, but may have felt foolish to act on something that may have been estimated to be a very unlikely possibili ty, and may have then reported positive estimations of their partner. If this is the case, then a couple of situational factors ma y have caused participants to decide it a better choice to play it safe. First, the negative interpreta tion of the ambiguous feedback may not have been seen as negative eno ugh to risk the consequences of retaliating. In other words, although the negative interpretation may have been id entified, the participants may not have seen this negative connotation as so strong that it deserved a response. A second possible factor is that instead of seeing the upcoming interaction and creative intelligence assessment as so important that negative communications should not be ri sked, is that perhaps the participants actually perceived little valu e in their performance on the upcoming assessment and so were not motivated to defend themselves as much as they would if the assessment were

PAGE 91

91 highly important. Although this second option se ems possible given the sometimes apathetic nature of participants in genera l, this does not seem to be the most parsimonious explanation as some participants were visibly upset by the feedb ack they received and cl early took it personally (as reported anecdotally by research assistants se rving as experimenters). Moreover, participants reported the measure to be a generally valid assessment of creative intelligenceeven the negative feedback participants (who saw the test as least valid) on average reported that they perceived the test to be a moderately valid measur e of creative intelligence with scores near the midpoint of a 9-point scale. Another possibility for the gene ral lack of differences betw een the positive feedback and ambiguous feedback conditions can be drawn from the literature on ingratia tion. This literature indicates that recipients of i ngratiation attempts are generally quite willing to believe that the ingratiation directed at them is sincere. Howeve r, witnesses to ingratiation attempts who are not themselves the target of the ingratiation attempt can readily identify the ingratiators true motive (Jones, 1964; Vonk, 2002). This distinction betw een the person on the receiving end of an insincere compliment and an observer of such an act may elucidate the general failure to find differences between responses to positive and ambiguous feedback. Specifically, in the current study, participants receiving ambiguous feedback may have reacted th e same way that a target of ingratiation responds: accept the insi ncere compliment at face value. In contrast, the participants in the pilot study evaluated the in terpretations of the feedback knowing that the feedback was not from a real person in response to their actual work. Therefore, pe rsonal biases that come from being the true recipient of an insincere co mpliment may not have come into play. People who find themselves as the target of ingratiation may not interpret insincere compliments accurately for several reasons. Most obviously, participants may be motivated to

PAGE 92

92 interpret it as sincere (Jones, 1964). Not only doe s this provide a boost to ones self-worth (Vonk, 2002), but it is also simpler to interact with a person who is sincere rather than insincere, and people may be motivated to avoid believing that they interact with insincere others. Additionally, Vonk (2002) proposed that people actively engaged in interaction with another person would have fewer cognitive resources ava ilable to second-guess a persons motives than someone who is not actively engaged in conversa tion and having to monitor, plan, and evaluate their own contribution to the interaction. In short, witnesses are able to focus more attention on the potential insincerity of an ingratiator/covert aggressor and thus more likely to detect insincerity. Although his sole test of this pos sibility did not receive strong support, Vonk assessed situations in which the consequences fo r insincerity are not as dire (i.e., a potential criticism and an upcoming evaluativ e interaction), as in the current study. In contrast, with the heightened consequences of the communicati on in the current study, the cognitive business explanation may still be a viable explanation for the current data. Alternatively, Vonk (2002) also assessed the pos sibility that a hei ghtened ingratiation self-concept discrepancy may result in a further c onsideration and questioni ng of the ingratiators motives and sincerity. Vonk assessed this possib ility via self-esteem and found that low selfesteem recipients of flattery believed ingratiato rs comments to be less accurate than high selfesteem flattery recipients. Vonk concluded that these low self-esteem participants experienced a greater discrepancy between their self-view and th e feedback they received from the ingratiators and were therefore more motivated to seriously evaluate the sincerit y of the feedback. In contrast, the high self-esteem participants di d not experience the same self-viewfeedback discrepancy and so they were not motivated to reassess the sincerit y of the feedback.

PAGE 93

93 This alternative vantage point explanation is in some ways illustrated via the discrepancy in how ambiguous feedback particip ants viewed the dual nature of the feedback they received and the messages they sent to thei r partner. If these participants did not see the dua l nature of the mixed messages, they would not have avoided these statements. Howeve r, participants who presumably saw little insincerity in the evalua tion of their own work significantly preferred positive messages to mixed messages. In other words, while these participants were oblivious to the insincerity of the messages they received, they were at least somewhat aware of the potential insincerity of the mixed messages they consid ered sending. Otherwise, these messages should have been endorsed as often as the positive messagesthey were not. One possibility for future research would be to examine the extent to which witnesses to such covert aggression alter their interaction behavior toward the vi ctim or the perpetrator of the aggression. Even if the covert aggression goes un noticed by the intended victim, the aggression may still have consequences for the victim or the aggressor. For example, as previous research has suggested (LaFontana & Cillessen, 2002; Sa lmivalli, Kaukiainen, & Lagerspetz, 2000), indirect and relational aggression are predictive of peer ac ceptance and social status. This increase in social status should parallel a similar increase in re spect (i.e., just as social status is boosted as a results of aggression, respect should also be increased after the use of aggression). Previous research does not specify whether this boon in social status is echoed in the victim of the aggression. After any given instance of aggr ession, a persons liking fo r their aggressor is almost certainly diminished, but the victim ma y also come to respect the aggressor more. However, given the dual nature of covert aggres sion and the requisite savoir faire necessary to properly encode the dual meaning, it is quite possible that indivi duals choosing to engage in covert aggression recogni ze that the targets of such aggression may never realize that they have

PAGE 94

94 been insulted. At times, it may even be plausible th at covert aggression is intended solely for the purpose of impressing witnesses to such events. As such, covert aggressors goal would not be simply to cause psychological pain in their victim, but rather the purpose is much more personalto impress witnesses (who should be more capable of determining the insincerity of the covert aggression) with the aggressors own intellect, communication skill, and willingness to attack another. The awareness of the social ramifications of covert aggression does not preclude that occasionally, people may engage in covert aggressi on without witnesses to the event or with the assumption that their target will not understand the intended slight. In fact, they may find the gullibility of the victim all that more entertaining. In short, people may engage in covert aggression simply for their own personal pleasure. However, to the extent that covert aggression is intricately linked to the presen ce of third parties, then future research should examine how and to what extent the presence of a third party in th e same social fiber aff ects the rate of covert aggression and the effectiveness of it. The presence of others should increase the use of covert aggression over overt aggression not only because it portrays the aggressor as a sophisticated social creature, but also because the presence of others within th e same social network amplifies concerns with maintaining a c ool, sophisticated, non-brutish image. In contrast, if covert aggression is truly intended as a social tool, then the presence of a thir d party not relevant to ongoing social interactions should have little effect on the implantation of covert aggression. Nature of the Interpersonal Relationship In addition to the possib ility that covert aggr ession is primarily a social tool, future research should investigate and be able to demonstrate that changing the nature of the social situation should similarly affect use of covert aggression. For example, covert aggression should be a particularly successful tool for those who fi nd themselves in a situation that prohibits overt

PAGE 95

95 aggressionsuch as aggressing against a higher-status victim. As such, the covert to overt aggression ratio should be heightened as the appropriateness of engaging in overt aggression diminishes. For example, some cultures (e.g., southern culture) disc ourage open aggression (Cohen, Vandello, Puente, & Rantilla, 1999) and as a result, covert aggr ession should be even more likely. The existence these kinds of norms do not necessarily have to reflect larger cultural expectations, however. Such variations could al so be a reflection of a specific peer-groups expectations for social behavior. Just as some individuals would rather hear a criticism in a straightforward manner and others prefer sugarcoate d criticism, there is also variability in the degree to which certain frie ndship groups prefer open and honest communication versus aggression disguised as something sweet. Other factors that may affect the use of c overt aggression include not only the relative status of individuals or the norms for interacti ons, but also perceived s ecurity in a particular relationship. Covert aggression, as a social tool, is presumab ly more important in some relationships than others. Specifi cally, relationships vary in th e degree to which they can be characterized by the game-playing nature of covert aggression. Some relationships, such as the early stages of romantic relations hips are much more characterized by game playing, in that they involve a strategic approach and reaction to the other pers ons behavior. However, as relationships grow and mature the role that such power concerns play decreases. The characterization of a mature relationship marked by openness and concern for the other does not have room for covert aggression. In fact, the use of covert aggr ession in such relationships may have quite the adverse effect on the relationship partner and any witnesses to such behaviors. Engaging in covert aggression in truly intimate relationships is likely not only to be more

PAGE 96

96 difficult given that an intimate partner will be less easily dupe d, but the expectations for many relationships is to discuss concerns openly. Doi ng otherwise may indicate a mark of immaturity. Gender Another noteworthy pattern is the lack of i mp act that gender of participants made. A great deal of literature and a number of authors would argue that ge nder plays an important role in who aggresses and how. Popular social psycho logical opinions hold that females aggress verbally and indirectly whilst males aggress directly and physically (e.g., Galen & Underwood, 1997). If so, females in this study would ha ve responded more negatively than men after receiving negative feedback, and pe rhaps would also have been part icularly reactive to instances of covert aggression and the us e of covert aggression. However, this was not found to be the case. In fact, there were essentia lly no gender differences in use of overt or covert aggression, or in response to overt aggression, covert aggre ssion, or no aggression (positive messages). Although this may seem surprising, one interpre tation of these findings would actually be consistent with previous research regarding so cial network density (Green, Richardson, & Lago, 1996). This research suggests that proclivity to engage in physic al or direct aggression versus verbal, relational, social, or indi rect aggression, does not result necessarily from gender itself, but rather should be considered under resources av ailable to men and women. Specifically, women tend to find themselves more heavily enmeshed in complicated social networks; their networks are generally larger and individu als in these denser social netw orks tend to know other members in the womans social network. In other words, in womens network, there tend to be many social connections shared among members of the social network. In contrast, men tend to have smaller social networks and thes e networks tend to be more fragm ented with each individual in a mans social network being considerably less likely to know other members of that social network than in a womans social network. This greater social connect edness makes engaging in

PAGE 97

97 indirect/verbal/social/ relational/covert aggression not only easier for women than men, but also more efficacious. Men who choose to aggress usi ng this strategy not only have fewer resources with which to carryout the aggression (e.g., spread ing a rumor is harder when the rumor can only be spread to one person), but th is aggression also has less social impact (the resulting negative evaluation of the victim due to the deleterious rumor will harm the victim less when only one person knows the rumor rather than many). Therefore, choosi ng to engage in nonphysical aggression is a much wiser choice for female aggressors than male aggressors. Moreover, engaging in this kind of aggr ession may spare a great deal of resources (physical energy, potential bodily harm, etc. ), over physical aggression. This disparity in the efficiency of covert aggression as a function of social network density is one factor that leads men to avoid this kind of aggression. However, Green, Richardson, and Lago (1996) also demonstrated that men who have higher social network density demonstrate an increased proclivity to engage in indirect aggression (and thus possibly covert aggression). In other words, when the social environment of me n and women are equated, gender differences in indirect aggression are muted. In the current study, male and female participants found themselves in an identical social situation: they would soon be te mporarily particip ating with one other same-sex person they did not know. In effect, this held the social environments of men and women equal. Following from the results of Gree n et al. (1996), this provides men and women the same social network resources and should th erefore eliminate gender differences. As such, the lack of gender effect s is perfectly consistent with prior research. Of course, other possible explanations for the lack of gender differen ces could be a simple statistical anomaly. However given the prevalen ce of null findings involving gender, this possibility seems, at best, a weak argument. Alternatively, perhaps cu ltural influences on

PAGE 98

98 available aggressive methods have altered the manner in which males and females engage in aggression. Specifically, as this age group becomes increasingly more reliant on non-face-to-face communication as a way to maintain interper sonal ties (i.e., cell phones, instant messaging, online social network communities, etc.), perhaps individualsmales and females alikehave, as a generation, begun to shift away from more direct forms of aggression. With less reliance on face-to-face interaction, there is perhaps less opportunity for males to practice direct or physical aggression. As a result, they may develop si milar aggressive behaviors as females. Expectation of a Future Interaction Another direction that future research s hould explore is how a long-term relationship affects the nature of covert aggression. The current study attempted to investigate covert aggression within the context of a somewhat dependent relationship that participants could not readily avoid. Participants receiv ing negative feedback who did not anticipate a future interaction did endorse mixed messages more strongly than negative feedback participants no t expecting an interaction, and also gave a weaker endorsement of positive messages when they did not expect an interaction. However, the future interaction manipulati on did not have the hypothesized effect. It was predicted that pa rticipants facing an interaction would strongly endorse mixed messages, but participants not f acing an interaction would feel less social concern about the consequences of engaging in overt aggression an d would therefore abandon mixed messages for straightforward negative messages. However, not only were ambiguous f eedback participants unaffected by the future meeting manipulation, but negative feedback part icipants were actually more likely to use mixed messages when they did not anticipate a future interaction then when they were. One possibility for this patte rn supports the above conjectu re that participants were generally being extremely cautious in their responses to ambiguous feedback. Specifically, an

PAGE 99

99 imminent interaction shoul d provoke participants to be more cau tious than participants not facing an imminent interaction. Therefore, in the face of an obviously negative attack, negative feedback participants expecti ng an interaction were less cauti ous and more strongly endorsed negative messages than ambiguous feedback participants expec ting an interaction. However, negative feedback particip ants not expecting an interaction were not as motivated to be cautious and could therefore afford to be more negativ e, resulting in a stronge r endorsement of mixed messages. This finding is not inconsistent with the reasoning presente d in this paper that participants not anticipating a fu ture interaction will be less con cerned about the consequences of their behavior than those expecti ng an interaction. This interpretati on does of course assume that negative feedback participants were aggressive ly motivated and perceived mixed messages as conveying a negative interpretation. Pilot testing clearly indica tes that these messages were indeed perceived as having dual meanings, and the general eschewing of these messages by positive and ambiguous participants indicates that the duality of meani ngs did not go entirely unnoticed. Of course, these nega tive feedback participants not expecting an interaction were not recklessly abandoning all thoughts of social cons equences and caution: they did not strongly endorse outright negative messages any more than negative feedback part icipants expecting an interaction. Perhaps, despite instructions that some part icipants would not work on the second creative intelligence assessment with their partner, partic ipants believed that they would in fact be meeting their partner at the end of the session or for some other tas k. Indeed, although the manipulation was generally successful, it was not a perfect manipulation a nd some participants who were told they would not be completing the second assessment with their partner still reported that they would be meeting their partner. This could have weakened the effect of this

PAGE 100

100 manipulation, making participants who should not have expected a meeting to be more cautious in the messages they sent. Another possibility is that the nature of the future interaction did not carry the psychological weight that is necessary to engage real life patterns of covert aggression. Perhaps the prospect of meeting another undergraduate wi th whom one would interact only briefly was not sufficient motivation for participants to be con cerned with the others tr ue intentions or worth stooping to aggression (covert or overt). Future research shou ld aim to increase the psychological magnitude of the future meeting, participants desire to maintain positive interpersonal relationships, and the consequences of not aggressing (e.g., lowered sociometric status, decreased respect within a group, etc.). Ideally, such research would include investigation of ongoing interpersonal groups. Unfortunately, the logist ics of such a project make this a difficult task to achieve. For example, previous unpub lished research (Miller & Schlenker, 2008), attempted to study covert aggression in actual frie ndship pairs. In order to study these friendship pairs, participants were asked to bring a friend to the lab. The resulting attempts to provoke participants to covertly aggress against their fr iends were unsuccessful an d essentially backfired when participants with the most motivation to ag gress against their friends being the least likely to do so. Norms for appropriate friendship beha vior were apparently heightened when the opportunity to break those norms presented th emselves. Moreover, friends who agree to participate in a psychology study are likely clos e enough that it would take very extraordinary circumstances to provoke true aggression against each othercircumstances that are not readily reproduced in a psychological laboratory. Perhaps experimental psychological research would be most successful in demonstrating covert aggression between mere acquaintances. In such situations, motivation to maintain a non-

PAGE 101

101 aggressive image might be particularly high. A nother alternative methodology to investigate the nature of covert aggression would be to create real experimental groups who would meet as part of an ongoing longitudinal study. As such, partic ipants would develop ties to the other participants, be concerned with their image in the group, and be accountable to the group for their behavior. Social Desirability It was pred icted that participants higher in social desirability w ould be less likely to endorse either negative message or mixed messages than low social desirability participants, but if they did, they would show a preference for mixed messages over negative messages. In the wake of negative feedback, part icipants with mixed messages av ailable behaved quite in line with the traditional nature of soci al desirability and this hypothesis: higher social desirability participants were less negative and more positive in their messages than low social desirability participants. However when participants could only send neutral messages instead of mixed messages, negative feedback participants re sponded quite differently. When high social desirability participants could not send mixed messages, they were significantly more negative than their low social desirability counterparts. In other words, when thei r hands were figuratively tied and they were forced to be direct with their negative messages, hi gh social desirability participants exhibite d more aggression. In some ways, this may seem counterintuitive. These high social desi rability participants are most concerned with maintaining a positive interpersonal image, and strongly endorsing negative messages does not seem to be the most effective method of achieving this goal. A couple of possibilities ex ist that may explain this patter n. One possibility is that these participants were more bothered by the negati ve feedback than low social desirability participants. As a result, the high social desirability participants perceived the feedback as more

PAGE 102

102 deserving of a negative response than low social desirability participants However, although this possibility seems plausible, partic ipants high versus low in social desirability did not differ in their emotional responses or views of th e partner as a function of feedback. An alternative account suggests that perhaps individuals high in social desirability regularly avoid straightforward negative messages by way of creating in th eir daily interactions ways to convey their own mixed messages. Le ft to their own devices, then, high social desirability participants may se emingly exhibit a lower rate of obvious aggression while they are in fact finding their own way to still be aggressive. Research on catharsis theory would suggest that by engaging in aggression (albeit socially presentable aggression), people will be more likely to aggress in the future. Most of the time, this increased aggression in high social desirability individuals will stil l be expressed in a more covert manner. However, when high social desirability individuals do not have the luxury of creating a socially acceptable way to express their aggression, they may actually exhi bit higher rates of open aggression than those individuals who are not so creative in their aggressive endeavors. Perspective Taking Based on previous research regarding the role of perspective taking on successful aggression, it was expected that participants hi ghest in perspective taki ng ability would be the most likely to engage in cove rt aggression by strongly endorsing m ixed messages. However, this was not the case as perspective taking did not pr edict rankings of mixed messages or neutral messages. While this finding does not support the original hypotheses regarding use of mixed messages, perspective taking was predictive of negative message endorsement. In keeping with the general pattern, participants receiving nega tive feedback endorsed negative messages more strongly when neutral messages were available than when mixed messages were available. However, this preference for neutral messages ov er mixed messages is pa rticularly evident in

PAGE 103

103 high perspective takers. In essence, high perspect ive takers were indeed more aggressive via negative messages when they did not have the ability to engage in mixed messages. In contrast, when mixed messages were available, high perspectiv e takers were actually less aggressive than their low perspective taking counterparts. One possible interpretation of this is that high perspective takers may exhibit more overt aggression but when other options are available, high perspective takers are not more aggressive. Given that negative messages in this study parall el closely indirect aggression (not face-to-face aggression that does not possess dual meanings), and traditional indi rect aggression literature has not allowed for the possibility of mixed messages, this is consistent with previous research finding that high perspective take rs can be more indirectly aggressive. Much of the existing literature regarding indirect a ggression and perspective taking has focused on younger children and teenagers. If covert aggre ssion is a more crafted and honed fo rm of aggression, then perhaps children and teenagers essentially do not have co vert aggression available to themnot because of situational constraints, but rather because they have not le arned how to successfully employ such a tactic. In contrast, more socially sophi sticated young adults may have learned the art of employing mixed messages in their aggression and favor this over more straightforward forms of aggression, and as such, perspective taking leads to an inverse relations hip in straightforward aggression. Distinguishing Covert Aggression from Other Forms of A ggression One question that can be raised regarding th ese findings is the exte nt to which covert aggression differs from indirect aggression or, more broadly, a ggression in general. Covert aggression shares many features with indirect ag gression in the sense that both have an element of concealment to them. Indirect aggression does not try to disguise the aggression, but rather simply eliminates the opportunity for victims to immediately defend th emselves. In contrast,

PAGE 104

104 covert aggression may be done faceto-face, and while the victims of covert aggression have the opportunity to aggress in retaliation if they so choose, the vi ctims must first discern whether the incident was actually malevolent. As a result victims may postpone any immediate response not because of inability (as is the case with indirect aggression), but because further consideration may be warranted. This similarity may result in future resear ch ignoring the distinction between the two. However, this could be a mistake. Covert aggression is a unique cons truct that introduces different interpersonal costs and benefits for t hose choosing to engage in it. For example, if asked to defend ones aggressive behavior, an in direct aggressors only defense is a lack of witnesses, which is something not always availa ble to the indirect aggressorsimply because the victim was not present for the behavior does not mean that others were not present. In contrast, the covert aggressors defense lies in whether the beha vior can truly be considered aggression or not. The covert aggressor would not deny the behavior occurred, but rather deny the aggressive interpretation was intended. In other words, there is no perfectly decisive indicator that the covert aggressor inte nded aggression. Future research should address the different interpersonal consequences of engaging in cove rt aggression. At any single moment in time, victims or witnesses may give the covert ag gressor the benefit of the doubt and apply a nonaggressive attribution to such behavior, bu t the kernel of doubt will be magnified with repeated exposure to a covert aggressors shif ty explanations for his/her behavior and will eventually come to view this person as an aggr essive individual, despite no single aggressive act having enough evidence to support that conclusion. In many ways this is similar to the effectiveness of excuses: alt hough excuses for poor performance do indeed work in the moment to avert negative attributions about skill level, they negatively a ffect the holistic assessment of a

PAGE 105

105 persons reliability and trustwor thiness. Similarly, covert aggr essors will come to develop reputations that precede them and warn pot ential victims to avoi d provoking the covert aggressor.

PAGE 106

106 APPENDIX A PERSONALITY PACKET (PART 1) Carefully read each sections ins tructions. Questions 1-33: Listed below are statements con cerning your attitudes, tr aits, and tendencies. Read each item and decide whether the statemen t is true or false as it pertains to you. 1 = True 2 = False 1. Before voting I thoroughly investigate th e qualifications of all the candidates. 2. I never hesitate to go out of my way to help someone in trouble. 3. It is sometimes hard for me to go on w ith my work if I am not encouraged. 4. I have never intensely disliked anyone. 5. On occasion I have had doubts about my ability to succeed in life. 6. I sometimes feel resentful when I don't get my way. 7. I am always careful about my manner of dress. 8. My table manners at home are as good as when I eat out in a restaurant. 9. If I could get into a movie without paying and be sure I was not seen, I would probably do it. 10. On a few occasions, I have given up doing someth ing because I thought too little of my ability. 11. I like to gossip at times. 12. There have been times when I felt like rebell ing against people in authority even though I knew they were right. 13. No matter who I'm talking to, I'm always a good listener. 14. I can remember "playing sick to get out of something. 15. There have been occasions when I took advantage of someone. 16. I'm always willing to admit it when I make a mistake. 17. I always try to prac tice what I preach. 18. I don't find it particularly difficult to get along with loud-mouthed, obnoxious people. 19. I sometimes try to get even, rather than forgive and forget. 20. When I don't know something I don't at all mind admitting it. 21. I am always courteous, even to people who are disagreeable. 22. At times I have really insist ed on having things my own way. 23. There have been occasions when I felt like smashing things. 24. I would never think of letting someone else be punished for my wrongdoings. 25. I never resent being asked to return a favor. 26. I have never been irked when people expr essed ideas very different from my own. 27. I never make a long trip without checking the safety of my car. 28. There have been times when I was quite jealous of the good fortune of others. 29. I have almost never felt the urge to tell someone off. 30. I am sometimes irritated by people who ask favors of me. 31. I have never felt that I was punished without cause. 32. I sometimes think when people have a misfor tune they only got what they deserved. 33. I have never deliberately said something that hurt someone's feelings.

PAGE 107

107 Questions 34-62: Using the 5 point scale show n below, indicate how uncharacteristic or characteristic each of the following statements is in describing you. 1 = extremely uncharacteristic of me 2 = somewhat uncharacteristic of me 3 = neither uncharacteristic or characteristic of me 4 = somewhat characteristic of me 5 = extremely characteristic of me 34. Some of my friends think I am hotheaded 35. If I have to resort to violence to protect my rights, I will. 36. When people are especially nice to me, I wonder what they want. 37. I tell my friends openly when I disagree with them. 38. I have become so mad I have broken thinks 39. I cant help getting into arguments when people disagree with me. 40. I wonder why sometimes I feel so bitter about things. 41. Once in a while, I cant control the ur ge to strike another person. 42. I am an even-tempered person. 43. I am suspicious of overly friendly strangers. 44. I have threatened people I know. 45. I flare up quickly but get over it quickly 46. Given enough provocation, I may hit another person. 47. When people annoy me, I may tell them what I think of them. 48. I am sometimes eaten with jealousy. 49. I can think of no good reason to ever hitting a person. 50. At times, I feel I have gotten a raw deal out of life. 51. I have trouble controlling my temper. 52. When frustrated, I let my irritation show. 53. I sometimes feel that people are laughing at me behind my back. 54. I often find myself disagreei ng with people. 55. If somebody hits me, I hit back. 56. I sometimes feel like a powder keg ready to explode 57. Other people always seem to get the breaks. 58. There are people who pushed me so far that we came to blows. 59. I know that friends talk about me behind my back. 60. My friends say that Im somewh at argumentative. 61. Sometimes I fly off the handle for no good reason. 62. I get into fights a little more than the average person.

PAGE 108

108 Questions 63-88: Here are a number of characteristics that may or may not apply to you. For example, do you agree that you are someone who likes to spend time with others? Please write a number next to each statement to indicate the extent to which you agree or disagree with that statement. Disagree Disagree Neither agree Agree Agree strongly a little nor disagree a little strongly 1 2 3 4 5 I See Myself as Someone Who... 63. Is talkative 64. Does a thorough job 65. Is original, comes up with new ideas 66. Is reserved 67. Can be somewhat careless 68. Is curious about many different things 69. Is full of energy 70. Is a reliable worker 71. Is ingenious, a deep thinker 72. Generates a lot of enthusiasm 73. Tends to be disorganized 74. Has an active imagination 75. Tends to be quiet 76. Tends to be lazy 77. Is inventive 78. Perseveres until the task is finished 79. Values artistic, aesthetic experiences 80. Is sometimes shy, inhibited 81. Does things efficiently 82. Prefers work that is routine 83. Is outgoing sociable 84. Makes plans and follows through with them 85. Likes to reflect, play with ideas 86. Has few artistic interests 87. Is easily distracted 88. Is sophisticated in art, music, or literature

PAGE 109

109 Questions 89-116: How frequently in the last 6 months might you engage in the behaviors listed below when you are angry with someone? Use the scale below. 1 2 3 4 5 Never Very Often 89. Yelled or screamed at them 90. Did things to irritate them 91. threatened to hit or throw something at them 92. made up stories to get them in trouble 93. did not show that I was angry 94. cursed at them 95. threw something at them 96. tried to make them look stupid 97. stomped out of the room 98. made negative comments about their appearance to someone else 99. hit (or tried to hit) them with something hard 100. insulted them or called them names to their face 101. talked the matter over 102. spread rumors about them 103. sulked and refused to talk about it 104. kicked (or tried to kick) the other person 105. dropped the matter entirely 106. took something that belonged to them 107. hit (or tried to hit) the othe r person but not with anything 108. gossiped about them behind their back 109. pushed, grabbed, or shoved them 110. called them names behind their back 111. told others not to associate with them 112. waited until I calmed down and th en discussed the problem 113. told others about the matter 114. threw something (but not at the other) or smashed something 115. destroyed or damaged something that belonged to them 116. gathered other friends to my side

PAGE 110

110 APPENDIX B PERSONALITY PACKET (PART 2) The following statem ents inquire about your thoughts and feelings in a variety of situations. For each item, indicate how well it describes you by c hoosing the appropriate number on the scale at the top of the page. Answer as honestly as you can. Thank you. 1 2 3 4 5 Does not Describes describe me very me well well 1. I often put off until tomorrow things I know I should do today. 2. Before criticizing somebody, I try to imagine how I would feel if I were in their place. 3. I am basically a lazy person. 4. I find satisfaction in working as well as I can. 5. I prefer to work in situations that require a high level of skill. 6. I more often attempt tasks that I am not sure I can do than tasks I know I can do. 7. If I'm sure I'm right about something, I don't waste much time listening to other people's arguments. 8. If Im not good at something, I would rather keep st ruggling to master it than move on to something I may be good at. 9. I find satisfaction in exceeding my previous pe rformance even if I dont outperform others. 10. I like to work hard. 11. I sometimes try to understand my friends better by imagining how things look for their perspective. 12. I believe that there are two sides to every question and try to look at them both. 13. I sometimes find it difficult to see things from the "other guy's" point of view. 14. Hard work is something I like to avoid. 15. I try to look at everybody's side of a disagreement before I make a decision. 16. I hate to see bad workmanship. 17. When I'm upset at someone, I usually try to "put myself in his shoes" for a while. 18. I can easily sit for a long time doing nothing. 19. I must admit I often do as little work as I can get away with. 20. I easily get bored if I dont have something to do. 21. When I see someone being taken advantage of, I feel kind of protective toward them. 22. I get a sense of satisfaction out of being able to say I have done a very good job on a project. 23. I would rather lean easy fun games than difficult thought games. 24. When I see someone being treated unfairly, I some times don't feel very much pity for them. 25. I often have tender, concerned feeli ngs for people less fortunate than me. 26. It is no use playing a game when you are playing with someone as good as yourself. 27. I would describe myself as a pretty soft-hearted person. 28. Sometimes I don't feel very sorry for other people when they are having problems. 29. Other people's misfortunes do not usually disturb me a great deal. 30. There is satisfaction in a job well done. 31. I am often quite touched by things that I see happen. 32. Part of the satisfaction in d oing something comes from seeing how good the finished product looks. 33. I would rather do something at which I feel c onfident and relaxed than something which is challenging and difficult. 34. I like to be busy all the time. 35. I feel like giving up quickly when things go wrong.

PAGE 111

111 APPENDIX C BASIC INFORMATION QUESTIONNAIRE F irst Name: Taylor Gender: Male Female Age: 21 Why are you at the University of Florida and what are your post-graduation plans? I am a senior here at UF. I came to the Univ ersity of Florida because of its good academic reputation at a pretty good price. I hope to attend graduate school next year after I graduate in May. Im taking intro psych right now with Dr. M iller and I really like it so far. Im thinking about taking another psycho logy class next semester. What do you like to do for fun? I enjoy hanging out with my fr iends during the weekend.

PAGE 112

112 APPENDIX D FEEDBACK Evaluator First Name: Taylor Form: Squirrel-Proof Birdfeeder GPACPL Form B-2.7 Time-limit: 6 minutes Please take a moment to review the problem assi gned to your fellow participant. Make sure that you fully understand what was described in the instructions. If you have any questions, please ask the researcher now. Once you understand the problem assigned to your fellow participant, ta ke a few moments to assess the solution developed by him/her in response to this problem. In the space below, please give your impressi on of this part icipants solution. Refer to the instructions associated with the participants sp ecific form of the test to determine how well his/her solution solves the problem. Positive Feedback: This design is a pretty creative one. I think it has a lot of potential to solve the squirrel proof birdfeeder in a useful and i nventive way. Overall, good job. This person probably has a lot of creative intelligence. Negative Feedback: This design is pretty bad. I dont think it would really work well, if at all. I can easily think of several different solutions that would work much better. This person probably doesnt have much creative intelligence. Ambiguous Feedback: This persons abilities are self evident. A thorough analysis of his/her performance will surprise you. This persons worth to his/her future empl oyer can only be imagined. This person has relatively few minor weaknesses. There is no doubt as to the ex tent of this persons creative intelligence.

PAGE 113

113 APPENDIX E FIRST IMPRESSIONS QUESTIONNAIRE (MIXED MESSAGES CONDITION) ***Conditio n: Mixed Messages. M denotes a mixed messages statement, N denotes a negative message statement and P denotes a positive message statement. In real world groups, people work with individuals about whom they have varying degrees of knowledge, ranging from barely knowing some to knowing others well. In all cases, the impressions people have of others can be important. Some impressions may be held with confidence whereas others may be little more than guesses or hunches. Depending on the condition to which youve been assigned, you may have exchanged more or less information with the other, and this may also affect your impressions and the confidence you have in them. Moreover, people are often aware of what others think of them before ever meeting them, so we need to take this into account. This questionnaire assesses your impressions. In th e condition to which you have been assigned, your impressions will be shown to the other person prior to the second assessment. However, you will not learn the other persons impressions of you until the end of the session. Please give us your impression of the other person, even if you are only guessing. Therefore, please review the statements below that describe individu als in a variety of domainsinterpersonal, task performance, etc. From the list, please select 8 statemen ts that you believe best describe the other person by placing an X on the 1st blank line next to the statement. Then, rank each of the ten selected statements from 1 to 8 where 1 = most descriptive, 2 = 2nd most descriptive, 3 = 3rd most descriptive, etc., through 8= 8th most descriptive by writing a number 1-8 on th e second blank line next to the statement. The statements you did not select should all be less descriptive of the person than the eight that you selected. Select Rank by writing selected an X statements ____ ____ The quality of this pe rsons work is well-known. M ____ ____ Several people actively dislike this person. N ____ ____ This person is someone I would like to get to know better. P ____ ____ This is not the kind of person you encounter everyday. M ____ ____ This person is not easy to respect. N ____ ____ This person is capable of doing many things. P ____ ____ This person is quite di stinct from his/her peers. M ____ ____ This person possesses many positive qualities. P ____ ____ This person is hard to get to know. N ____ ____ This person probably has a well-known reputation. M ____ ____ A lot of people would like to be this persons close friend. P ____ ____ This person finds novel tasks difficult. N ____ ____ This person is a well-rounded individual. P ____ ____ This person has made quite an impression on me. M ____ ____ Few people consider this pe rson one of their close friends. N ____ ____ Many people come to this pe rson to seek personal advice. P ____ ____ This person is quite interesting. M ____ ____ People usually find this person to have an unpleasant personality. N

PAGE 114

114 APPENDIX F FIRST IMPRESSIONS QUESTIONNAIRE (NE UTRAL MESSAGES CONDITION) ***Condition: Neutral Messages NL denotes a neutral message statement, N denotes a negative message statement and P denotes a positive message statement.*** In real world groups, people work with individuals about whom they have varying degrees of knowledge, ranging from barely knowing some to knowing others well. In all cases, the impressions people have of others can be important. Some impressions may be held with confidence whereas others may be little more than guesses or hunches. Depending on the condition to which youve been assigned, you may have exchanged more or less information with the other, and this may also affect your impressions and the confidence you have in them. Moreover, people are often aware of what others think of them before ever meeting them, so we need to take this into account. This questionnaire assesses your impressions. In th e condition to which you have been assigned, your impressions will be shown to the other person prior to the second assessment. However, you will not learn the other persons impressions of you until the end of the session. Please give us your impression of the other person, even if you are only guessing. Therefore, please review the statements below that describe individu als in a variety of domainsinterpersonal, task performance, etc. From the list, please select 8 statemen ts that you believe best describe the other person by placing an X on the 1st blank line next to the statement. Then, rank each of the ten selected statements from 1 to 8 where 1 = most descriptive, 2 = 2nd most descriptive, 3 = 3rd most descriptive, etc., through 8= 8th most descriptive by writing a number 1-8 on th e second blank line next to the statement. The statements you did not select should all be less descriptive of the person than the eight that you selected. Select Rank by writing selected an X statements ____ ____ This person has taken a formal logic course. NL ____ ____ Several people actively dislike this person. N ____ ____ This person is someone I would like to get to know better. P ____ ____ This person probably has pets. NL ____ ____ This person is not easy to respect. N ____ ____ This person is capable of doing many things. P ____ ____ He/she lives off campus. NL ____ ____ This person possesses many positive qualities. P ____ ____ This person is hard to get to know. N ____ ____ This person comes from a large family. NL ____ ____ A lot of people would like to be this persons close friend. P ____ ____ This person finds novel tasks difficult. N ____ ____ This person is a well-rounded individual. P ____ ____ He/she wears corrective lenses. NL ____ ____ Few people consider this pe rson one of their close friends. N ____ ____ Many people come to this person to seek personal advice. P ____ ____ This person is originally from Florida. NL ____ ____ People usually find this person to have an unpleasant personality. N

PAGE 115

115 APPENDIX G PRELIMINARY INTERACTION ASSESSMENT Instructions. This scale consists of a number of words that de scribe different feelings and emotions. Read each item and then mark the appropr iate answer in the space next to that word. Indicate to what extent you feel this way right now, at this mo ment. Use the following scale to record your answers: 1.2.3.4.5 very slightly a little moderately quite a bit extremely or not at all It is very important that you answ er as accurately as possible to ensure the validity of the final assessment. Your responses to these questions will not be shown to the other participant. _____ interested _____ irritable _____ distressed _____ alert _____ excited _____ ashamed _____ upset _____ inspired _____ strong _____ nervous _____ guilty _____ determined _____ scared _____ attentive _____ hostile _____ jittery _____ enthusiastic _____ active _____ proud _____ afraid

PAGE 116

116 Please answer the following questions by circling th e number that best corresponds to your opinion. Depending on the condition to which you have been assigned, some of th e questions may seem irrelevant because you dont expect to meet your partner. If you are assigned to such a condition, imagine how a meeting with your partner would be. It is very important that you answ er as accurately as possible to ensure the validity of the final assessment. Your responses to these questions will not be shown to the other participant. 1. Do you believe that you will like the other participant? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 2. Do you think that the other participant is generally friendly? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 3. Do you believe you will respec t the other participant? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 4. How much do you think you will enjoy work ing with the other participant? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 5. How well do you think this person gets along with other people? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely well well well well well 6. Do you believe that the othe r participant will like you? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 7. How much do you think the othe r participant respects you? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit

PAGE 117

117 8. How much creative intelligence do you th ink the other participant possesses? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely creatively creatively creativ ely creatively creatively intelligent intelligent intelligent intelligent intelligent 9. How much do you value the other participants opinion regard ing creative intelligence problem solving? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 10. How much creative intelligen ce do you believe you possess? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 11. How much creative intell igence does the other par ticipant believe you have? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely creatively creatively creativ ely creatively creatively intelligent intelligent intelligent intelligent intelligent 12. How much creative intelligence do you think th e other participant believes he/she has? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely creatively creatively creativ ely creatively creatively intelligent intelligent intelligent intelligent intelligent 13. How well do you think the first assessment of creative intelligen ce actually assessed creative intelligence? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely well well well well well 14. How well do you think the second assessment of creative intelligence will assess your creative intelligence? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely well well well well well

PAGE 118

118 15. How clearly do you express your thoughts and feelings to others? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely Well well well well well 16. How well do you understand the thoughts and f eelings that others express to you? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly extremely Well well well well well 17. To what extent do you prefer working with others versus working alone? 1..2..3. 4..5..6..7..8..9 always or mostly with others mostly always or Almost always alone and alone with almost always alone equally others with others 18. To what extent do you think your final answer on the second assessment will be equal parts your work and the ot her participants work? 1..2..3. 4..5..6..7..8..9 entirely mostly e qually mostly entirely other other me and me me person person other person 19. To what extent do you think the se cond assessment will be difficult? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 20. How positively did you evaluate the other persons work? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly Extremely at all very positive Positively Positively Positively positively 21. How positively did the other person evaluate your work? 1..2..3. 4..5..6..7..8..9 not at all not very so mewhat fairly Extremely at all very positive Positively Positively Positively positively

PAGE 119

119 22. To what extent do you think that the other person is straightforward? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 23. To what extent do you think that the other person is sincere? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 24. To what extent do you think that the ot her person is a skilled communicator? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit 25. How much are you worried about inte racting with the other person? 1..2..3. 4..5..6..7..8..9 not not somewhat a good extremely at all much bit

PAGE 120

120 APPENDIX H STATEMENT SELECTION MEANS Table H-1. Statement selection means Feedback Statement type Mixed messages Neutral messages Ambiguous feedback Alternative messages 3.09ab (0.70) 2.57c (0.83) Negative messages 0.07de (0.33) 0.03df (0.16) Positive messages 4.84g (0.77) 5.41h (0.86) Negative feedback Alternative messages 3.34aij (1.01) 3.31i (0.90) Negative messages 1.83 (1.51) 2.83 (1.76) Positive messages 2.83j (1.93) 1.86 (1.52) Positive feedback Alternative messages 2.79bk (0.80) 2.79ck (0.82) Negative messages 0.05el (0.22) 0.05fl (0.05) Positive messages 5.15gm (0.87) 5.16hm (0.90) Values reflect average number of messages se lected out of a possible 6. Orthogonal simple effects comparisons are presented in the table, with means having at leas t one factor in common and sharing a subscript not significantly different from one another. Standard deviations are indicated in parentheses.

PAGE 121

121 APPENDIX I STATEMENT SELECTION MEANS WITH OFFSET COMPARISON GROUPS NOT EXPECTING A FUTURE INTERACTION Table I-1. Statem ent selection means with offs et comparison groups not expecting a future interaction Feedback Statement type Future meeting No future meeting Ambiguous feedback Mixed messages 3.09ab (0.70) 2.97a (0.81) Negative messages 0.07c (0.33) 0.11c (0.32) Positive messages 4.84d (0.77) 4.92d (0.73) Negative feedback Mixed messages 3.34bef (1.01) 3.74e (1.12) Negative messages 1.83g (1.51) 2.10g (1.27) Positive messages 2.83fh (1.93) 2.16h (1.37) Values reflect average number of messages selected out of a possible 6. Orthogonal simple effects comparisons are presented in the table, with means having at leas t one factor in common and sharing a subscript not significantly different from one another. Standard deviations are indicated in parentheses.

PAGE 122

122 LIST OF REFERENCES Archer, J., & Coyne, S. M. (2005) An integrated review of i ndirect, relational, and social aggression. Personality and Social P sychology Review, 9 (3) 212-230. Archer, J., & Webb, I. A. (2006). The relation between scores on the Buss-Perry Aggression Questionnaire and aggressive acts, impulsi veness, competitiveness, dominance, and sexual jealousy. Aggressive Behavior 32, 464-473. Baron, R., & Richardson, D. R. (1994). Human Aggression. New York: Plenum. Berglas, S., & Jones, E. E. (1978). Drug choice as a self-han dicapping strategy in response to noncontingent success. Journal of Personality and Social Psychology, 36 (4) 405-417. Bjrkqvist, K., Lagerspetz, K ., Kaukiainen, A. (1992). Do gi rls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18, 117-127. Bjrkqvist, K., sterman, K., & Lagerspetz, K. M. J. (1994) Sex differences in covert aggression among adults. Aggressive Behavior, 20, 27-33. Bjrkqvist, K., sterman, K., Lagerspetz, K., La ndau, S. F., Caprara, G., & Fraczek, A. (2001). Aggression, victimization and sociometric stat us: Findings from Finla nd, Israel, Italy and Poland. In J. M. Rameriez & D. S. Richardson (Eds.), Crosscultural approaches to aggression and reconciliation (pp. 111). Huntington, NY: Nova Science. Bradbury, T.N., & Fincham, F.D. (1992). Attri butions and behavior in marital interaction. Journal of personality and Social Psychology, 63, 613-628. Bureau of Justice Statistics Crime Charac teristics (2007). Retrieved April 2, 2007, from http://www.ojp.usdoj.gov/bjs/cvict_c.htm #relate Bushman, B. J., & Anderson, C. A. (2001). Is it time to pull the plug on hostile versus instrumental aggression dichotomy? Psychological Review, 108 (1) 273-279. Buss, A. H. (1961). The Psychology of Aggression. New York: John Wiley and Sons, Inc. Buss, A. H., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and Social Psychology, 63 (3), 452-459. Cairns, R. B., Cairns, B., Neckerman, H., Ferguson, L., & Garipy, J. (1989). Growth and aggression: 1. Childhood to early adolescence. Developmental Psychology, 25 320. Campos, B., Keltner, D., Beck, J.M., Gonzaga, G. C., & John, O.P. (2007). Culture and teasing: The relational benefits of reduced desire for positive self-differentiation. Personality and Social Psychology Bulletin, 33 (1) 3-16.

PAGE 123

123 Cohen, D., Vandello, J., Puente, S., & Rantilla, A. (1999). When you call me that, smile! How norms for politeness, interaction styles, and a ggression work together in southern culture. Social Psychology Quarterly, 62 (3), 257-275. Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression / correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum. Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social-information processing mechanisms in childrens social adjustment. Psychological Bulletin, 115 (1) 74-101. Crick, N. R., & Grotpeter, J. (1995). Relationa l aggression, gender, and social-psychological adjustment. Child Development, 66, 710. Crick, N. R., Werner, N., Casas, J., OBrien, K., Nelson, D., Grotpeter, J., et al. (1999). Childhood aggression and gender: A new look at an old problem. Nebraska Symposium on Motivation, 45, 75. Crowne, D. P., & Marlowe, D. (1960). A new s cale of social desirabi lity independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10 85. Gaertner, S. L. & Dovidio, J. F. (1977). The s ubtlety of white racism arousal, and helping behavior. Journal of Personality and Social Psychology, 35 (10) 691-707. Galen, B. R., & Underwood, M. K. (1997). A developmental investigation of social aggression among children. Developmental Psychology, 33, 589. Geen, R. G., Stonner, D., & Shope, G. L. (1975) The facilitation of a ggression by aggression: Evidence against the catharsis hypothesis. Journal of Personality and Social Psychology, 31(4) ,721-726. Green, L. R., Richardson, D. R., & Lago, T. (1996). How do friendship, indirect, and direct aggression relate? Aggressive Behavior 22, 81-86. Goldstein, S. E., Tisak, M. S., Persson, A. V., & Boxer, P. (2006). Childrens evaluations of ambiguous provocation by relationa lly aggressive, physically aggressive, and prosocial peers. British Journal of Developmental Psychology, 24 (4) 701-708. Gordon, R. A. (1996) Impact of ingratiation on judgments and evaluations: A meta-analytic investigation. Journal of Personality and Social Psychology, 71 (1) 54-70. Hardin, C. D., & Higgins, E. T. (1996). Shared reality: How social verification makes the subjective objective. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition: Vol 3. The interpersonal context, (pp. 28-84). New York: Guilford Press.

PAGE 124

124 Jones, E. E. (1964). Ingratiation. New York: Appleton-Century-Crofts. Kaukiainen, A., Bjrkqvist, K., La gerspetz, K., sterman, K., Salmivalli, C., Rothberg, S., & Ahlbom, A. (1999). The relationships between social intelligence, empathy, and three types of aggression. Aggressive Behavior, 25, 81-89. Klar, Y., & Giladi, E. E. (1997). No one in my group can be below the groups average: A robust positivity bias in favor of anonymous peers. Journal of Personality and Social Psychology, 73(5) 885-901. Kowalski, R. M., Howerton, E., & McKenzie, M. (2001). Permitted disrespect: Teasing in interpersonal interactions. In R. M. Kowalksi (Ed.), Behaving Badly: Aversive Behaviors in Interpersonal Relationships (pp. 177-202), Washington, DC: American Psychological Association. LaFontana, K. M., & Cillessen, A. H. N. ( 2002). Childrens percepti on of popular and unpopular peers: A multimethod assessment. Developmental Psychology, 38 (5) 635-647. Lagerspetz, K., Bjrkqvist, K., & Peltonen, T. (1988) Is indirect aggressi on typical of females? Gender differences in aggressiveness in 11to 12-year old children. Aggressive Behavior 14, 403-414. Miller, M. L. & Schlenker, B. R. (2008). [D isguising aggression between friends]. Unpublished raw data. Millon, T., Davis, R., & Millon, C. (1997) MCMI-III: Millon Clinical Multiaxial Inventory-III Manual 2nd ed. Minneapolis, MN: NCS Pearson, Inc. Richardson, D. R., & Green, L. R. (2003). Defi ning direct and indire ct aggression: The Richardson Conflict Response Questionnaire. Revue Internationale de Psychologie Sociale, 16 (2) 11-30. Salmivalli, C., Kaukiainen, A., & Lagerspetz, K. (2000). Aggression and sociometric status among peers: Do gender and type of aggression matter? Scandinavian Journal of Psychology, 41, 17-24. Schlenker, B. R. (1980). Impression management: The self-c oncept, social identity, and interpersonal relations Monterey, CA: Brooks/Cole. Schlenker, B. R. (2003). Self-presentation. In M. R. Leary and J. P. Tangney (Eds.), Handbook of self and identity (pp. 492-518). New York: Guilford. Schlenker, B. R., & Britt, T. W. (1999). Be neficial impression management: Strategically controlling information to help friends. Journal of Personality and Social Psychology, 76, 559-573.

PAGE 125

125 Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in as sessing rater reliability. Psychological Bulletin, 86 (2) 420-428. Snyder, M. L., Kleck, R. E., Strenta, A., & Ment zer, S. J. (1979). Avoidance of the Handicapped: An attributional ambiguity analysis. Journal of Personality and Social Psychology, 37 (12) 2297-2306. Tedeschi, J. T., & Felson, R. B. (1994). Violence, Aggression, and Coercive Actions. Washington, DC: American Psychological Association. Vonk, R. (2002). Self-serving interpretations of flattery: Why ingratiation works. Journal of Personality and Social Psychology, 82 (4) 515-526. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS Scales. Journal of Personality and Social Psychology, 54, 1063-1070.

PAGE 126

126 BIOGRAPHICAL SKETCH Marisa L. Miller was born in Valdosta, Geor gia in 1981. She grew up in various cities in Georgia, graduating from Troup County Com prehensive High School in LaGrange, Georgia in 1999. She graduated summa cum laude with honors from Mercer University in Macon, Georgia in 2003 with a Bachelor of Ar ts in psychology and dual minors in art and photography. In fall 2003, she began her graduate work at the University of Florida were she earned her Master of Science in psychology, before beginning her doctoral work in social psychology.