O objetivo geral desta tese, composta por um artigo teórico e três empíricos, foi investigar o desenvolvimento de atitudes raciais em crianças. O primeiro estudo foi uma revisão sistemática de estudos da Psicologia brasileira sobre preconceito racial. O segundo envolveu a criação e validação de um banco de estímulos para utilização em pesquisas científicas, o BICMulticor. O terceiro utilizou medidas implícitas (Priming Avaliativo e Teste de Associação Implícita) e explícitas de atitude para avaliar o desenvolvimento de atitudes raciais em crianças brancas, pardas e pretas de Porto Alegre e Salvador. Participaram 542 crianças, de seis a 14 anos de idade. Foram encontradas diferenças nas atitudes implícitas entre os grupos raciais e também entre as duas cidades avaliadas. O quarto foi um estudo sobre as diferenças de categorização racial entre baianos, gaúchos e norte-americanos. A categorização racial sofreu forte influência contextual e foi semelhante na Bahia e nos Estados Unidos. / This dissertation, composed by one theoretical and three empirical papers, aimed to investigate the development of race attitudes in children. The first study was a systematic review of Brazilian psychological studies on racial prejudice. The second one involved the development and validation of a set of children’s portraits to be used as stimuli on scientific research, the BIC-Multicor. The third one used implicit (Evaluative Priming and the Implicit Association Test) and explicit measures of attitude to investigate the development of race attitudes in White, Pardo and Black children (N = 542), aged six to 14, from Porto Alegre and Salvador. Implicit attitudes were different between cities and also between racial groups. The forth study assessed differences in racial categorization in Bahia, Rio Grande do Sul and the United States. Racial categorization was influenced by context and similar results were found between participants from Bahia and the US.

The objectives of this research were to create explicit and implicit measures of prejudice against homeless, and to identify factors associated with negative attitudes toward this population. To meet these objectives several studies collected in four articles were made. In the first article the literature on prejudice against homeless was reviewed. In this study was found that both, in Latin America and in the countries of industrialized West, have been reported stereotypes towards this population related to mental illness and drug addiction, as well as acts of discrimination related to limitations in the use of public space. Article two corresponded to a study of cross-cultural creation and validation of a scale of prejudice against homeless, created with basics of Item Response Theory and the Allport's scale of prejudice. The created scale presented evidence of construct validity, convergent validity and good internal consistency. It was also verified, in part, the Allport's theory of the scale of prejudice. The third article aimed at identifying factors associated with prejudice against homeless, and to build and test a hierarchical model to explain variations in the levels of this prejudice. The proposed model had four levels of prejudice mediations, good fit indices, and included the variables social dominance orientation, basic values, egocentrism, perception of insecurity and quality of contact with homeless people. In the fourth article was described the process of creating an implicit association test about attitudes toward homeless, and was tested scores correlation levels acquired with this test and those acquired with the scale established in article two. The results indicated that there are strong associations between concepts representing the homeless people and negative stereotypes towards this population. No correlation between the scores on the two measures was found, in line with other studies on implicit and explicit measures of stereotypes. / Los objetivos de esta tesis fueron la creación de medidas explícitas e implícitas del prejuicio contra habitantes de calle, y la identificación de algunos factores asociados con actitudes negativas hacia esta población. Para responder a estos objetivos se realizaron varios estudios reunidos en cuatro artículos y presentados en este documento como capítulos. En el primer capítulo, se realizó un estudio de revisión de la literatura sobre el prejuicio contra habitantes de calle. En este estudio se encontró que tanto en Latinoamérica como en los países de occidente industrializado, se han reportado estereotipos hacia esta población relacionados con la enfermedad mental y la adicción a las drogas; así como actos de discriminación relacionados con limitaciones en el uso del espacio público. El capítulo dos corresponde a un estudio de creación y validación transcultural de una escala de prejuicio contra habitantes de calle, fundamentada en la teoría de la respuesta al ítem y en la teoría de la escalada comportamental del prejuicio de Allport. La escala creada presentó evidencias de validez de constructo, validez convergente y buenos índices de consistencia interna. También se comprobó parcialmente la hipótesis de la escala del prejuicio de Allport. En el tercer capítulo se buscó identificar los factores asociados al prejuicio contra habitantes de calle, así como construir y probar un modelo jerárquico para explicar variaciones en los niveles de este prejuicio. El modelo propuesto tuvo cuatro niveles de mediación del prejuicio, buenos índices de ajuste, e incluyó las variables dominancia social, valores básicos, egocentrismo, percepción de inseguridad y calidad del contacto con habitantes de calle. En el cuarto capítulo se describió el proceso de creación de un Test de Asociaciones Implícitas sobre actitudes hacia habitantes de calle, y se probó los niveles de correlación entre los puntajes adquiridos con este test y los adquiridos con la escala creada en el artículo dos. Los resultados indicaron que hubo fuertes asociaciones entre los conceptos que representan a los habitantes de calle y los estereotipos negativos hacia esta población. No se encontró una correlación entre los puntajes en las dos pruebas, acorde con los resultados de otros estudios sobre medidas implícitas y explícitas de estereotipos.

The objectives of this research were to create explicit and implicit measures of prejudice against homeless, and to identify factors associated with negative attitudes toward this population. To meet these objectives several studies collected in four articles were made. In the first article the literature on prejudice against homeless was reviewed. In this study was found that both, in Latin America and in the countries of industrialized West, have been reported stereotypes towards this population related to mental illness and drug addiction, as well as acts of discrimination related to limitations in the use of public space. Article two corresponded to a study of cross-cultural creation and validation of a scale of prejudice against homeless, created with basics of Item Response Theory and the Allport's scale of prejudice. The created scale presented evidence of construct validity, convergent validity and good internal consistency. It was also verified, in part, the Allport's theory of the scale of prejudice. The third article aimed at identifying factors associated with prejudice against homeless, and to build and test a hierarchical model to explain variations in the levels of this prejudice. The proposed model had four levels of prejudice mediations, good fit indices, and included the variables social dominance orientation, basic values, egocentrism, perception of insecurity and quality of contact with homeless people. In the fourth article was described the process of creating an implicit association test about attitudes toward homeless, and was tested scores correlation levels acquired with this test and those acquired with the scale established in article two. The results indicated that there are strong associations between concepts representing the homeless people and negative stereotypes towards this population. No correlation between the scores on the two measures was found, in line with other studies on implicit and explicit measures of stereotypes. / Los objetivos de esta tesis fueron la creación de medidas explícitas e implícitas del prejuicio contra habitantes de calle, y la identificación de algunos factores asociados con actitudes negativas hacia esta población. Para responder a estos objetivos se realizaron varios estudios reunidos en cuatro artículos y presentados en este documento como capítulos. En el primer capítulo, se realizó un estudio de revisión de la literatura sobre el prejuicio contra habitantes de calle. En este estudio se encontró que tanto en Latinoamérica como en los países de occidente industrializado, se han reportado estereotipos hacia esta población relacionados con la enfermedad mental y la adicción a las drogas; así como actos de discriminación relacionados con limitaciones en el uso del espacio público. El capítulo dos corresponde a un estudio de creación y validación transcultural de una escala de prejuicio contra habitantes de calle, fundamentada en la teoría de la respuesta al ítem y en la teoría de la escalada comportamental del prejuicio de Allport. La escala creada presentó evidencias de validez de constructo, validez convergente y buenos índices de consistencia interna. También se comprobó parcialmente la hipótesis de la escala del prejuicio de Allport. En el tercer capítulo se buscó identificar los factores asociados al prejuicio contra habitantes de calle, así como construir y probar un modelo jerárquico para explicar variaciones en los niveles de este prejuicio. El modelo propuesto tuvo cuatro niveles de mediación del prejuicio, buenos índices de ajuste, e incluyó las variables dominancia social, valores básicos, egocentrismo, percepción de inseguridad y calidad del contacto con habitantes de calle. En el cuarto capítulo se describió el proceso de creación de un Test de Asociaciones Implícitas sobre actitudes hacia habitantes de calle, y se probó los niveles de correlación entre los puntajes adquiridos con este test y los adquiridos con la escala creada en el artículo dos. Los resultados indicaron que hubo fuertes asociaciones entre los conceptos que representan a los habitantes de calle y los estereotipos negativos hacia esta población. No se encontró una correlación entre los puntajes en las dos pruebas, acorde con los resultados de otros estudios sobre medidas implícitas y explícitas de estereotipos.

O objetivo geral desta tese, composta por um artigo teórico e três empíricos, foi investigar o desenvolvimento de atitudes raciais em crianças. O primeiro estudo foi uma revisão sistemática de estudos da Psicologia brasileira sobre preconceito racial. O segundo envolveu a criação e validação de um banco de estímulos para utilização em pesquisas científicas, o BICMulticor. O terceiro utilizou medidas implícitas (Priming Avaliativo e Teste de Associação Implícita) e explícitas de atitude para avaliar o desenvolvimento de atitudes raciais em crianças brancas, pardas e pretas de Porto Alegre e Salvador. Participaram 542 crianças, de seis a 14 anos de idade. Foram encontradas diferenças nas atitudes implícitas entre os grupos raciais e também entre as duas cidades avaliadas. O quarto foi um estudo sobre as diferenças de categorização racial entre baianos, gaúchos e norte-americanos. A categorização racial sofreu forte influência contextual e foi semelhante na Bahia e nos Estados Unidos. / This dissertation, composed by one theoretical and three empirical papers, aimed to investigate the development of race attitudes in children. The first study was a systematic review of Brazilian psychological studies on racial prejudice. The second one involved the development and validation of a set of children’s portraits to be used as stimuli on scientific research, the BIC-Multicor. The third one used implicit (Evaluative Priming and the Implicit Association Test) and explicit measures of attitude to investigate the development of race attitudes in White, Pardo and Black children (N = 542), aged six to 14, from Porto Alegre and Salvador. Implicit attitudes were different between cities and also between racial groups. The forth study assessed differences in racial categorization in Bahia, Rio Grande do Sul and the United States. Racial categorization was influenced by context and similar results were found between participants from Bahia and the US.

Stigmatized obese individuals tend to internalize the pervasive weight stigma which might lead to self-discrimination and increased psychopathology. While explicit and implicit weight stigma can be measured using self-report questionnaires and Implicit Association Tests (IAT), respectively, the assessment of self-discrimination relied solely on self-report. The present study sought to develop an IAT measuring implicit self-discrimination (SD-IAT) in samples of obese individuals with and without binge-eating disorder (BED). Seventy-eight individuals were recruited from the community and individually matched in three groups. Obese participants with BED, obese participants without BED (OB), and a normal weight control group without eating disorder psychopathology (HC) were assessed with the SD-IAT and other measures relevant for convergent and discriminant validation. Results revealed significantly higher implicit self-discrimination in the BED group when compared to both OB and HC. Furthermore, significant correlations were found between the SD-IAT with body mass index, experiences of weight stigma, depressive symptoms, and implicit self-esteem. Finally, implicit self-discrimination predicted eating disorder psychopathology over and above group membership, and experiences of weight stigma. This study provides first evidence of the validity of the SD-IAT. Assessing implicit self-discrimination might further increase understanding of weight stigma and its significance for psychosocial functioning among vulnerable obese individuals.

Unverfälschbarkeit stellt ein wichtiges Gütekriterium psychologischer Testverfahren dar. Dieses Kriterium gilt dann als erfüllt, wenn das Testverfahren auf Grund seiner Konstruktion keine Steuerung oder Verzerrung der Ausprägung von Testwerten seitens der Versuchspersonen ermöglicht (vgl. Moosbrugger & Kelava, 2012). Im Gegensatz zu direkten Verfahren (z.B. Fragebogen und Interviews), bei welchen die Ausprägung hinsichtlich eines Merkmales durch Selbstbeschreibung der Versuchspersonen erfragt wird und eine Verfälschung (z.B. durch sozial erwünschtes Antwortverhalten) nicht ausgeschlossen werden kann, wurde indirekten Verfahren (z.B. dem Impliziten Assoziationstest; IAT; Greenwald, McGhee, & Schwartz, 1998) lange Zeit Immunität gegen Fälschungsversuche unterstellt. Diese begründet sich unter anderem durch die Annahme, dass mittels indirekter Verfahren implizite Merkmale gemessen werden. Implizite Merkmale unterscheiden sich von den „eher klassischen“ expliziten Merkmalen, welche vorwiegend mittels direkter Verfahren gemessen werden. Ein wesentlicher Unterschied besteht darin, dass Versuchspersonen nicht notwendigerweise um die Ausprägung hinsichtlich ihrer impliziten Merkmale wissen und dass sie diese Ausprägung auch nicht kontrollieren können (vgl. De Houwer, 2006; De Houwer & Moors, 2007, in press). Die theoretischen Annahmen bezüglich der Eigenschaften impliziter Merkmale bzw. Messergebnisse legen zwei Implikationen nahe. Erstens: Wir können implizite Merkmale ausschließlich über indirekte Zugänge erfassen, da diese nicht notwendigerweise bewusst sind und so eine Selbstauskunft nicht möglich erscheint. Zweitens: Personen können ihre impliziten Messergebnisse nicht kontrollieren und folglich auch nicht verfälschen. Vermutlich gab es auch aus diesem Grund vor wenigen Jahren einen regelrechten Boom, der zu der Entwicklung einer Vielzahl indirekter Verfahren zur Erfassung impliziter Merkmale geführt hat. Ob jedoch die Messergebnisse dieser Verfahren tatsächlich implizit und damit nicht verfälschbar sind, darf nicht nur theoretisch unterstellt, sondern muss empirisch überprüft werden (vgl. De Houwer, 2006). Der IAT gilt als das bekannteste, reliabelste und valideste indirekte Verfahren (Bosson, Swan, & Pennebaker, 2000; Rudolph, Schröder-Abé, Schütz, Gregg, & Sedikides, 2008). In meiner Dissertation habe ich mich aus diesem Grund der empirischen Überprüfung auf Verfälschbarkeit des IATs gewidmet. Die vorliegende Dissertation besteht aus insgesamt fünf Kapiteln. Das 1. Kapitel bildet eine theoretische Einführung zu den Themen Fälschung im diagnostischen Kontext und zum IAT. Grundlegende Befunde und Fragen zur Verfälschbarkeit des IATs werden dargestellt. Kapitel 2 bis 4 bilden empirische Beiträge meiner Forschung, die sich jeweils schwerpunktmäßig mit unterschiedlichen Aspekten der Verfälschbarkeit des IATs beschäftigen. In Kapitel 2 wird der Frage nachgegangen, unter welchen Bedingungen der IAT verfälschbar ist. Bis dato haben die wenigen existierenden Studien ein sehr widersprüchliches Bild bezüglich der Verfälschbarkeit des IATs aufgezeigt. Ein Grund hierfür könnte sein, dass potentiell relevante Faktoren, welche die Verfälschbarkeit des Verfahrens beeinflussen können, noch nie gemeinsam in einer Studie untersucht wurden. Die vorliegende Studie wurde genau mit diesem Ziel konstruiert und durchgeführt. Die Ergebnisse verweisen auf ein komplexes Zusammenspiel verschiedener Faktoren und zeigen auf, unter welchen Bedingungen der IAT verfälschbar ist. Implikationen dieser Ergebnisse werden kritisch diskutiert. In Kapitel 3 werden die Fragen beantwortet, wie Personen den IAT verfälschen und ob Fälschung im IAT detektierbar ist. Die Forschung hat sich bislang nur bedingt damit beschäftigt, was fälschende Personen tun, um ihre Messergebnisse wie gewünscht zu beeinflussen. Es wurde auch noch nicht untersucht, ob Versuchspersonen unter verschiedenen Bedingungen (z.B. Fälschungsziel: hohe vs. niedrige Testwerte) unterschiedliche Strategien anwenden. Dennoch wurden Indices vorgeschlagen, welche in der Lage sein sollen, Fälschung im IAT zu detektieren (Agosta, Ghirardi, Zogmaister, Castiello, & Sartori, 2011; Cvencek, Greenwald, Brown, Gray, & Snowden, 2010). In der vorgestellten Studie habe ich einerseits untersucht, welche Strategien fälschende Personen anwenden und ob sie, je nach Bedingung, zu unterschiedlichen Strategien greifen. Andererseits habe ich untersucht, welche dieser Strategien tatsächlich mit erfolgreicher Fälschung des IATs einhergehen. Schließlich habe ich untersucht, ob die in der Vergangenheit vorgeschlagenen Indices tatsächlich in der Lage sind, erfolgreiche FälscherInnen zu detektieren. Meine Ergebnisse zeigen, dass fälschende Personen unterschiedliche Strategien anwenden, um ihr Ziel zu erreichen. Damit verbunden zeigte sich auch, dass es schwerer ist als bislang angenommen, erfolgreiche FälscherInnen im IAT zu detektieren. Implikationen dieser Ergebnisse werden kritisch diskutiert. Kapitel 4 beschäftigt sich mit der Frage, ob kognitive Fähigkeiten ein erfolgreiches Fälschen im IAT erleichtern. Bisher wurden diese Fähigkeiten nur mit Fälschungserfolg in direkten Verfahren in Verbindung gebracht (vgl. Hartshorne & May, 1928; Nguyen, Biderman, & McDaniel, 2005; Ones, Viswesvaran, & Reiss, 1996; Pauls & Crost, 2005; Snell, Sydell, & Lueke, 1999; Tett, Freund, Christiansen, Fox, & Coaster, 2012; Weiner & Gibson, 2000). In der vorgestellten Studie habe ich untersucht, ob sie auch beim Fälschen des IATs eine Rolle spielen. Besonders habe ich mich dabei für die Rolle des g Faktors der Intelligenz, der Verarbeitungsgeschwindigkeit und der Konzentrationsfähigkeit interessiert. Die Ergebnisse meiner Studie zeigen auf, dass einige dieser Prädiktoren tatsächlich einen Einfluss auf den Fälschungserfolg im IAT haben. Implikationen dieser Ergebnisse werden kritisch diskutiert. Das 5. Kapitel bildet eine Zusammenführung und Integration der Befunde meiner Forschung in die bestehende Theorie. Zudem werden ein Ausblick für die weitere Forschung sowie Empfehlungen für die Praxis gegeben.

Evolutionäre Herangehensweisen hatten in der Persönlichkeitspsychologie nicht den gleichen Erfolg wie in anderen Bereichen der Psychologie. In dieser Dissertation werden zwei alternative evolutionäre Herangehensweisen an die Persönlichkeitspsychologie diskutiert und angewendet. Die evolutionsgenetische Herangehensweise fragt, warum genetische Varianz in Persönlichkeitsunterschieden existiert. Im ersten Teil dieser Dissertation werden verschiedene evolutionsgenetische Mechanismen, die genetische Varianz erklären können, verglichen. Auf Grundlage evolutionsgenetischer Theorie und empirischen Befunden aus der Verhaltensgenetik und Persönlichkeitspsychologie wird geschlussfolgert, dass ein Mutations-Selektions-Gleichgewicht genetische Varianz in Intelligenz gut erklären kann, während ausgleichende Selektion durch Umweltheterogenität die plausibelste Erklärung für genetische Unterschiede in Persönlichkeitseigenschaften ist. Komplementär zur evolutionsgenetischen Herangehensweise beginnt die „Life History“-Herangehensweise damit, wie Menschen ihre Ressourcen in evolutionär relevante Lebensbereiche investieren. Im zweiten Teil der Dissertationsschrift wird diese Herangehensweise am Beispiel von Investitionsunterschieden in Langzeit- versus Kurzzeit-Paarungstaktiken (wie im Konstrukt der Soziosexualität abgebildet) erläutert. Zwei neue Maße zur Erfassung von Soziosexualitätskomponenten werden vorgestellt. Während das revidierte Soziosexuelle Orientierungsinventar (SOI-R) ein Fragebogen zur Erfassung der Facetten „Verhalten“, „Einstellung“ und „Begehren“ ist, wurde mit dem Single-Attribute Impliziten Assoziationstest (SA-IAT) eine neue Methode zur indirekten Erfassung impliziter Soziosexualität entwickelt. Beide Maße zeigten konkurrente Validität in Onlinestudien, aber nur der SOI-R erwies sich als prädiktiv für Paarungstaktiken, einschließlich beobachtetem Flirtverhalten sowie der Zahl der Sexualpartner und Veränderungen im Beziehungsstatus innerhalb der nächsten 12 Monate. / Evolutionary approaches have not been as successful in personality psychology as they were in other areas of psychology. In this thesis, two alternative evolutionary approaches to personality psychology are discussed and applied. The evolutionary genetic approach asks why genetic variance in personality differences exists. In the first part of this thesis, three evolutionary genetic mechanisms that could explain genetic variance in personality differences are assessed: selective neutrality, mutation-selection balance, and balancing selection. Based on evolutionary genetic theory and empirical results from behavior genetics and personality psychology, it is concluded that selective neutrality is largely irrelevant, that mutation-selection balance seems best at explaining genetic variance in intelligence, and that balancing selection by environmental heterogeneity seems best at explaining genetic variance in personality traits. Complementary to the evolutionary genetic approach, the life history approach starts with how people allocate their resources to evolutionarily relevant life tasks. In the second part of this thesis, differences in the allocation to long-term versus short-term mating tactics (as reflected in the construct of sociosexuality) are used as a case to exemplify this approach. Two new measures for the assessment of sociosexuality components are presented. While the revised Sociosexual Orientation Inventory (SOI-R) is a questionnaire that assesses the facets Behavior, Attitude and Desire, the sociosexuality Single-Attribute Implicit Association Test (SA-IAT) is a new methodic development aimed to assess implicit sociosexuality indirectly. Both measures showed concurrent validity in online studies, but only the SOI-R facets were predictive of mating tactics, including observed flirting behavior, as well as for the number of sexual partners and changes in romantic relationship status over the following 12 months.

This study applied the Process Dissociation Procedure (Bornstein, 2002) to test independence between personality processes represented by different implicit measurement techniques. In contrast to the commonly adopted literal view of dual processes in personality theory, the study predicted that two implicit measures (CRT-A and IAT-A) and one explicit measure (NEO-AH) of aggressive disposition would dissociate with each other in their 1) intercorrelations, 2) predictions of behavioral criteria of aggressiveness, and 3) potential moderation by situational cues. These hypotheses were generally, though not completely, supported. Most importantly, the two implicit measures dissociated in their lack of correlation and differential prediction of behavioral criteria, unaffected by changes in situational cues. As predicted, the CRT-A and the NEO-AH dissociated in their intercorrelations, predictions, and moderation by incentives. The IAT-A and the NEO-AH dissociated in their lack of intercorrelation and their differential moderation by changes in incentive conditions. As predicted, only the explicit measure was moderated by changes in incentive conditions. Unexpectedly, IAT-A and the NEO-AH were statistically indistinguishable in their prediction of behavioral criteria of aggression. The findings provided strong support for the hypotheses predicting multiple independent implicit personality processes.

To study of cognitive fear networks and associations, indirect experimental paradigms like the Implicit Association Test (IAT, Greenwald et al., 1998) or the Extrinsic Affective Simon Task (EAST, De Houwer, 2003) may be helpful, as they promise to assess the structure of specific associations using a performance based approach without having to ask the participant for a verbal report. Three studies investigated the usefulness and characteristics of IAT and EAST. The aim of the first study was to measure fear associations towards spiders among spider phobic and non-phobic participants as well as in a group of spider enthusiasts. Results indicate that the IAT paradigm is sensitive to the strength of fear relevant associations and able to predict anxious behavior beyond the predictions of direct measures such as questionnaires. The second study focused on some of the mechanisms that underlie IAT effects. With a newly developed masked IAT, these experiments investigated the influences of individual stimuli and superordinate categories on IAT performance. Besides theoretical implications, the results also provide practical, relevant applications for the use of IAT experiments. A third study applied the EAST to investigate how different context conditions lead to differential activation of cognitive schemata in fear of spiders. One can conclude that the impact of automatic threat associations depends on the activated context, and that the EAST is suitable for the assessment of fear associations and their current activation level. This dissertation leads to the conclusion that the performance based methodology of the IAT and EAST is a useful and practical approach to reflect fear associations in phobia indirectly. At this point, the use of indirect measures is still at its beginning, and requires intensive methodological and theoretical efforts. These paradigms, however, may become useful for possible implications in psychopathology and other fields of psychology.

Meta-analytic data highlight the value of the Implicit Association Test (IAT) as an indirect measure of personality. Based on evidence suggesting that confounding factors such as cognitive abilities contribute to the IAT effect, this study provides a first investigation of whether basic personality traits explain unwanted variance in the IAT. In a gender-balanced sample of 204 volunteers, the Big-Five dimensions were assessed via self-report, peer-report, and IAT. By means of structural equation modeling (SEM), latent Big-Five personality factors (based on self- and peer-report) were estimated and their predictive value for unwanted variance in the IAT was examined. In a first analysis, unwanted variance was defined in the sense of method-specific variance which may result from differences in task demands between the two IAT block conditions and which can be mirrored by the absolute size of the IAT effects. In a second analysis, unwanted variance was examined in a broader sense defined as those systematic variance components in the raw IAT scores that are not explained by the latent implicit personality factors. In contrast to the absolute IAT scores, this also considers biases associated with the direction of IAT effects (i.e., whether they are positive or negative in sign), biases that might result, for example, from the IAT's stimulus or category features. None of the explicit Big-Five factors was predictive for method-specific variance in the IATs (first analysis). However, when considering unwanted variance that goes beyond pure method-specific variance (second analysis), a substantial effect of neuroticism occurred that may have been driven by the affective valence of IAT attribute categories and the facilitated processing of negative stimuli, typically associated with neuroticism. The findings thus point to the necessity of using attribute category labels and stimuli of similar affective valence in personality IATs to avoid confounding due to recoding.