Association for Research in Personality

Symposia 1: Friday, July 17 10:00 a.m.-11:15a.m. Grand Ballroom

1. Personality in the Magnet: Using fMRI to Study Individual Differences

Colin G. DeYoung, University of Minnesota

Jeremy R. Gray, Yale University

Angus W. MacDonald, University of Minnesota

Brian W. Haas, Stony Brook University and Center for Interdisciplinary Brain Sciences Research, Stanford University School of Medicine

Neuroscience approaches to the study of personality are becoming increasingly common, as personality researchers attempt to coordinate psychological and biological understandings of personality traits. One of the most powerful neuroscience approaches is the use of functional magnetic resonance imaging (fMRI) to track blood flow corresponding to neural activity in the brain. fMRI provides a map of the activation of every part of the brain during a condition of interest, relative to a control condition. These activation maps can be examined in relation to individual difference measures, including personality assessments. This symposium presents four papers describing research on various personality traits using fMRI. The traits studied include intellect, extraversion, agreeableness, anxiety, and suspiciousness, and the tasks employed in the magnet include viewing emotion faces, using working memory, and making economic decisions. Collectively, these studies provide an excellent introduction to the variety of questions about personality that can be addressed using fMRI. Additionally, they highlight some of the many ways in which personality traits may be associated with brain function: Traits may moderate the effects of particular stimuli or emotion manipulations on brain activity; brain activity may mediate associations between traits and behavior; and trait differences between individuals may correspond to differences in brain activity. In addition to demonstrating that fMRI can provide useful data to test personality theory and to expand the nomological network surrounding personality traits, we hope to give our audience a sense of what is required to do good research on individual differences using fMRI, and we welcome questions. The symposium will close with an open panel discussion among symposium members and the audience, as time permits.

1.1. Intellect as distinct from Openness: Differences revealed by fMRI of working memory

Colin G. DeYoung,University of Minnesota

Noah A. Shamosh, Yale University

Adam E. Green,Yale University

Todd S. Braver, Washington University

Jeremy R. Gray, Yale University

Relatively little is known about the neural bases of the Big Five personality trait Openness/Intellect. This trait is composed of two related but separable aspects, Openness to Experience and Intellect. On the basis of previous behavioral research (DeYoung, Peterson, & Higgins, 2005), we hypothesized that brain activity supporting working memory would be related to Intellect but not Openness. To test this hypothesis we used functional magnetic resonance imaging (fMRI) to scan a sample of 104 healthy adults, as they performed a difficult working memory task, using both words and faces as stimuli in alternating blocks, contrasted with periods of rest. Intellect (and not Openness) was found to correlate with working memory performance and with performance-related brain activity, in left lateral anterior prefrontal cortex (aPFC) and posterior medial frontal cortex (pMFC). Neural activity in these regions mediated the association between Intellect and working memory performance, potentially implicating these regions in the neural substrate of Intellect. Previous research suggests that left lateral aPFC is involved in the integration of abstract information from multiple cognitive operations, whereas pMFC is involved in monitoring goal-directed performance. These functions are plausible psychological correlates of Intellect. Intellect was also correlated significantly with intelligence test scores, but the association of Intellect with brain activity could not be entirely explained by intelligence.

1.2. Anxiety, working memory, and processing efficiency: Insights from fMRI

Jeremy R. Gray, Yale University

Christina Fales, Deanna M. Barch, and Todd S. Braver, Washington University, St. Louis

According to the processing-efficiency hypothesis (Eysenck et al., 2007), anxious individuals are thought to require greater activation of brain systems supporting cognitive control (e.g., dorsolateral prefrontal cortex; DLPFC) in order to maintain equivalent performance to nonanxious subjects. A recent theory of cognitive control (Braver et al., 2007) has proposed that reduced cognitive efficiency might occur as a result of changes in the temporal dynamics of DLPFC recruitment. In this study, we used a mixed blocked/event-related fMRI design to track transient and sustained activity in DLPFC while high and low anxious participants performed a working-memory task. The task was performed after viewing videos designed to induce neutral or anxiety-related moods. After the neutral video, the high-anxious participants had reduced sustained but increased transient activation in working-memory areas, compared to low-anxious participants. The high-anxious also showed extensive reductions in sustained activity of "default network" areas (possible deactivation). After the negative video, the low-anxiety group shifted their activation dynamics in cognitive control regions to resemble those of the high-anxious. These results suggest that reduced cognitive control in anxiety might be due to a transient rather than sustained pattern of working memory recruitment.

1.3. Role of Medial PFC Activation Differences in MZ Twins Discordant for Persecutory Personality Traits

Angus W. MacDonald, III, James N. Porter, and Melissa K. Johnson, University of Minnesota

Robert F. Krueger, Washington University of St. Louis

There is a strong neuroeconomics literature on social decision-making constructs such as trust. Trust can be said to occur when an agent allows another person to determine the agent’s gains or loses. Such paradigms use scenarios where the other player gains at the agent’s expense. However, individuals who feel persecuted report mistrust even when others have no incentive to make them lose. To investigate the functional neuroanatomy of persecutory personality traits, the current experiment used fMRI and an experimental condition where the other player had no monetary incentive to make the participant lose. There is evidence from patients suffering from persecutory ideation that this symptom dimension is not heritable. Therefore, we examined 24 monozygotic (MZ) non-psychiatric twin pairs selected to be concordant and discordant for self-reported levels of persecutory ideation, to explore brain regions that reflected the differential impact of the environment on brain functioning. Participants made a series of parallel decisions with varying potential for gains and losses. These involved either an impersonal partner (a coin flip, Risk Aversion condition) or an anonymous human partner. The partner had either a $5 incentive to make the participant lose (Rational Mistrust condition) or a $5 cost to make the participant lose (Suspiciousness condition). In-scanner behavioral performance conformed to predictions. Medial prefrontal cortex (mPFC) was differentially active during the Suspiciousness condition, and this difference predicted the extent to which twins were high in trait suspiciousness. Importantly, discordance analyses showed that the extent to which twins differed in self-reported suspiciousness corresponded with the extent to which they differed in mPFC activity. These results are consistent with models suggesting that mPFC plays a key role in theory of mind processing. The findings further suggest that the environmental factors that lead to differences between twins on persecutory traits map onto a difference in mPFC activation.

1.4. Personality and Automaticity: A case for Agreeableness and Extraversion using functional Magnetic Resonance Imaging

Brian W. Haas, Ph.D, Stony Brook University and Center for Interdisciplinary Brain Sciences Research, Stanford University School of Medicine

Turhan Canli, Ph.D, Stony Brook University and Graduate Program in Genetics, Stony Brook University

Individual differences in personality have been demonstrated to correspond with a wide range a cognitive, behavioral and affective tendencies. In order to accurately measure these tendencies, studies often utilize explicit instructions. Recently, with advancements in brain imaging techniques such as with functional Magnetic Resonance Imaging (fMRI) and with an understanding of cognitive/affective neuroscience we have begun to explore the automatic neural processes that have previously been speculated to be associated with personality. In a series of two studies, we provide fMRI evidence that the personality traits agreeableness and extraversion are associated with differential amount and duration of brain activation in response to affective stimuli without the use of explicit instructions of emotional evaluation. Here, we demonstrate that during the implicit processing of fearful facial expressions, higher scores of agreeableness are associated with greater right lateral prefrontal cortex activation, a region previously shown to be engaged during emotion regulation. Additionally, we demonstrate that during the implicit processing of happy facial expressions, higher scores of extraversion are associated with quicker rates of habituation within the rostral anterior cingulate cortex, a region previously shown to be engaged during affective appraisal. Combined, these studies provide a framework in which implicit or automatic tendencies believed to be associated with personality can be investigated. We anticipate that a more advanced understanding of automatic processing associated with personality will be cardinal in developing behavioral and neural models of the relative vulnerability to developing mood disorders such as anxiety and depression.

Symposia 2: Friday, July 17 10:00 a.m.-11:15a.m. Heritage Ballroom

2. Challenges in Cultural Comparisons of Personality Traits and Processes: When Can We Trust Self-Reports?

Co-Chairs: Laura P. Naumann and Oliver P. John; University of California, Berkeley

How should we interpret apparent cultural differences in personality traits and processes? For example, when studying personality across different cultural or ethnic groups, can we simply take cultural differences in self-reports (e.g., of conscientiousness or emotion regulation) at face value? These talks seek empirical evidence for the causes and mechanisms that explain actual and apparent cultural differences. First, Buchtel, Heine, and Norenzayan consider the finding that observer-perceptions of national character are not always correlated with mean self-ratings by members of these national groups. They argue that individuals in different countries use different-reference group standards when evaluating themselves; this self-perception process can then distort potential cultural differences. Using behavioral and demographic predictors of conscientiousness, they demonstrate that national character personality profiles may better capture true cultural differences in personality. Second, Nairán Ramirez-Esparza and colleagues report that Mexicans’ self-reported personality scores for Extraversion and Agreeableness are lower than expected from their actual behavior. Third, Naumann and John demonstrate that both cultural values and reference-group standards need to be taken into account as mediating mechanisms when interpreting Asian-White differences in self-reported Openness and Conscientiousness. Finally, Eng and John demonstrate that for emotional suppression, an emotion regulation strategy with observable consequences, Asian-White differences in self-reported use do align with the common belief that East-Asians exert more control over their emotions. As highlighted by discussant Oliver P. John, these four talks all caution that different processes may affect self-reports of personality, depending on the context and the personality domain being studied. However, self-reports should not be thrown out with the bathwater. Instead, researchers should use multi-method assessments of personality, including peer reports, behavioral assessments, and demographic predictors to verify the validity of cultural differences, allowing us to better understand when we should—and shouldn’t—take self-reported cultural differences in personality at face value.

2.1. Comparing Personality across Cultures

Emma E. Buchtel, Steven J. Heine, and Ara Norenzayan, University of British Colmbia

How should we measure average personality differences between cultures? Much research on cross-cultural differences relies on comparison of average self-reports. However, the comparison of average self-report data is plagued by significant methodological problems, such as the reference-group effect. In particular, when self-reporting their personality, participants may compare themselves to the implicit norms of their culture, such as the actual average of their group, the assumed average of the group, or ideal standards. These artifacts may lead to cross-cultural comparisons that show artificially reduced differences, or even differences in the opposite direction than expected. In a recent influential analysis of international personality data (Terracciano et al., 2005), it was found that there was no correlation between self-reports and perceptions of national character. Does this mean that perceptions of national character are groundless stereotypes, or does it mean that the average self-reports were an invalid way of comparing personal across cultures? Using behavioral and demographic predictors of conscientiousness, we found evidence that the perceptions of national character were a more valid measure of actual cross-cultural differences than other measures of personality, such as self-reports and peer-reports. The evidence suggests that careful consideration of methodological artifacts is needed when comparing self-reports across cultures.

2.2. Personality Differences Across Cultures: Unraveling the Paradox of Extraversion and Agreeableness

Nairán Ramirez-Esparza, Institute for Learning and Brain Sciences, University of Washington

Samuel D. Gosling, University of Texas at Austin

Matthias Mehl, University of Arizona

James W. Pennebaker, University of Texas at Austin

Stereotypes about Mexicans are that they are sociable and outgoing and that they are agreeable, friendly, and polite. However, in self-reports, Mexicans rate themselves as less extraverted and agreeable than Americans. What can account for these paradoxical findings? One possible answer is that people’s lay beliefs might not reflect real cross-cultural differences in personality (Terraciano et al., 2005). That is, perhaps Mexicans are in fact less sociable and agreeable than Americans, and folk beliefs are wrong. Another possibility is that individuals’ perceptions of how they typically behave (when they complete a self-report questionnaire) do not correspond very well to how they actually behave. Indeed, in two studies, we found that for Extraversion and Agreeableness behavioral personality does not match self-reported personality. For example, Mexicans saw themselves as less sociable than Americans, but they behaved more sociably in their everyday lives. We propose that cultural differences in self-presentational biases may be driving these paradoxical effects. Specifically, we suggest that Americans are showing a self-enhancement bias when responding to self-reports, especially when responding to highly socially desirable traits such as extraversion and agreeableness. Likewise, that Mexicans when responding to socially desirable traits are manifesting a modesty bias. This idea suggests that perhaps Americans’ and Mexicans’ self-views interact with aspects of cultural norms, such as self-enhancement and modesty. This research underscores the importance of using alternative methods to understand the intriguing paradoxes and puzzles so prevalent in cross-cultural personality research.

2.3. The Self-Critical Asian: Myth or Reality?

Laura P. Naumann and Oliver P. John, University of California, Berkeley

Research suggests that East Asians have a tendency to view themselves in more critical and less positive ways compared to Westerners who generally see themselves too positively (Heine et al., 1999). In a series of studies, we found that Asian Americans self-reported that they were less open (e.g., creative, unconventional) and less conscientious (e.g., reliable, self-disciplined) than Whites. Do these self-critical perceptions reflect an accurate self-assessment or biased self-perception? To test this, we examined if these personality differences could be explained by differences in (a) values, (b) reference-group standards, and (c) behavior. Asian Americans valued openness attributes less than Whites and this difference mediated cultural differences in openness. The same was not true for conscientiousness; both groups rated this domain as highly (and equally) important. Instead, Asian Americans perceived a much higher in-group standard for conscientiousness. An experimental follow-up demonstrated that both groups perceived themselves lower when compared to the high conscientious standard held for Asians. Finally, to move beyond self-reported personality, we examined whether these personality differences replicated in behavioral and peer reports. Consistent with these domain-specific explanations, Asian Americans, who valued openness less, performed fewer openness behaviors (e.g., trying something new) and peers rated them as less open. In the absence of a high reference-group comparison, Asian Americans did not differ from Whites in how many conscientious behaviors (e.g., working on class assignments) they performed or in peer ratings. Cultural differences in self-reported personality are not solely the result of Asian self-criticism. For openness, Asian Americans endorse more traditional values suggesting that differences in openness may be real (rather than self-critical views). In contrast, Asian Americans contrast themselves against the very high standard in-group standard in self-evaluations of conscientiousness, yet behave as conscientiously as Whites do, suggesting that self-perceptions are biased (i.e., self-criticism) rather than actual behavioral differences.

2.4. Emotional Control in East Asians and European Americans: More Complicated than Folk Conceptions Suggest

Joshua S. Eng and Oliver P. John, University of California, Berkeley

Common folk conceptions suggest East Asians exert more control over their emotions than European Americans do. Does this portrayal have empirical support? If so, are these differences due to acculturation (e.g., adopting interdependent or independent values) or other causes (e.g., genetic differences)? Study 1 examined self-reported individual differences in two aspects of emotion control or regulation—reappraisal and suppression—comparing 4 groups that differed in acculturation to US culture: European-Americans, East-Asian Americans (born in the US), East Asians (born in Asia) living in the US, and East Asians living in Asia. There were no differences for reappraisal; for suppression we found systematic acculturation effects, with greater exposure to US culture predicting less use of suppression. Moreover, even within the group of foreign-born East-Asians immigrants, longer exposure to US culture predicted less suppression use. To ensure that these suppression findings were not due to self-report artifacts, Study 2 replicated them with peer-reports. Study 3 examined whether cultural differences in suppression held across specific emotions and tested independent and interdependent self-construal as mediators. We found no cultural differences for negative emotions; however, as predicted, European Americans suppressed positive emotions (particularly pride) less frequently than did East Asians. These differences were mediated by independence but not interdependence. We propose that general folk beliefs and empirical evidence do converge in this case because individual differences in suppression are common, highly observable, and thus reliably reported by self and various kinds of observers. However, our findings suggest a more nuanced picture of cultural differences in emotion regulation than folk conceptions would lead us to believe: East Asians’ greater emotional control was limited to (a) a particular regulation strategy (suppression) and (b) one type of emotion valence (positive emotions), and (c) explained by the greater importance of independence (or authentic self-expression) in European-American than East-Asian culture.

Symposia 3: Friday, July 17 11:30 a.m.-12:45a.m. Grand Ballroom

3. Hierarchical models of personality: What’s up?

Ken Sheldon, University of Missouri (organizer)

Jack Mayer, University of New Hampshire

Jennifer Pals Lilgendahl, Haverford College

Brian Little, Harvard University

Hierarchically-organized models of personality arrange personality, its internal systems such as perception and cognition, and outer influences, such as society and culture, according to the degree they are molar (big), on the one hand, or molecular (small) on the other. These models appeal greatly to personality psychologists – if their number and use is any measure. Models that are roughly hierarchical in their form extend from Freud’s id, ego, and superego to Carver & Scheier’s (1981) hierarchical organization of action). In the last 15 years, new hierarchical models of the personality system have emerged which promise more contemporary integrations. Important issues for these new models include the proper number (or type) of levels to specify; whether the model is meant to predict behavior in real-time or to conceptually arrange important themes and concepts in personality theory; whether and how the models can be tested by empirical data; and what, exactly, should be located at the top of the personality hierarchy. This symposium will explain “what’s up” in hierarchical models of personality, both by updating listeners on contemporary research and by presenting recent thinking about the embedding of personality within higher levels of analysis and organization. Jennifer Pals Lilgendahl will first describe new developments in McAdams’ (1996) ground-breaking “three tiers” model of personality. Then Ken Sheldon will describe his SLOPIC (6 levels of personality in context) model, which somewhat reorganizes and extends the original McAdams model. Next, Jack Mayer will offer an alternative hierarchical arrangement, and comment on its difference from the first two models. Finally, Brian Little, long interested in hierarchical models himself, will serve as an animated discussant.

3.1. Taking a “Principled” Approach to Personality

Jennifer Pals Lilgendahl, Haverford College

In 1996, McAdams asked the simple question, “What do we know when we know a person?”, and offered an answer for personality psychologists that included three levels of personality description: dispositional traits, personal concerns (e.g., schemas, developmental challenges), and life narratives. The use of “level” was not so much a reference to a true hierarchical structure as it was a reference to level of contextualization, from broad, de-contextualized traits to the uniquely formed and fully contextualized life story. Since then, this model has evolved into what McAdams and Pals (2006) referred to as “a new big five” – five fundamental principles for an integrative science of personality that includes evolution and culture in addition to the original three levels. In this talk, I will begin by briefly reviewing the five principles and how they are expected to relate to one another to form a useful schematic for understanding the whole person. I will then use contemporary research examples to illuminate some of the unique qualities and uses of the model. One unique contribution of this model is its strong emphasis on the life story. For example, the model takes narrative seriously, seeing narrative not as epiphenomenal to level 1 and level 2 constructs but as a mode of memory/thought that is central to personality functioning. Additionally, the model provides a particularly rich lens through which to examine culture, as will be demonstrated through a discussion of current research on bicultural identity. Overall, the principles are broadly useful in that they provides a powerful organizing language not unlike the big five trait taxonomy that, when applied consistently, could help the field to proceed in a more integrated fashion.

3.2. The SLOPIC model: Six levels of personality in context

Ken Sheldon, University of Missouri

In this talk I will discuss my “Six levels of personality in context” (SLOPIC) model (Sheldon, 2004, 2006, 2008). The SLOPIC model attempts to enumerate and hierarchically locate social-personality influences upon human behavior, ranging from needs to traits to goals to selves to social relations to culture. The SLOPIC model assumes that each level of organization’s effects are irreducible to lower levels of organization, and thus that hierarchical pluralism will be necessary in any “final” theory. The model is based upon McAdams’ ground-breaking “three tiers” model (McAdams, 1996), but differs from that model by a) attempting to locate personality within an even broader reality framework acknowledging molecular, biological, brain, and cognitive processes, b) taking the three tiers of personality more seriously as emergent levels of organization within the person each of which builds off of the levels below, c) locating evolved human nature at the bottom tier and cultural relations at the top tier of the hierarchy, and d) providing heuristics for designing any number of integrative multi-level modeling studies addressing many levels of predictor simultaneously. Data supporting the utility of the model for integrative study design will be presented, specifically for the case of understanding psychological well-being. Finally, the model will be compared with McAdams and Pal’s (2006) “New Big Five” model, which provides its own updating of McAdams’ original model, with Carver and Scheier’s (1998) hierarchical model of personality, which focuses on the nested organization of goal-striving only, and with McCrae and Costa’s model of personality functioning, which reduces other aspects of personality functioning to trait processes.

3.3. A tale of two (plus more) models of personality

John D. Mayer, University of New Hampshire

Structural models of personality and its environment(s) are those that concern divisions that are long-term and stable. A few historical examples include divisions of the mind into the id, ego, and superego, or into motives, emotions, and cognitions (Freud, 1923; Mendelssohn, 1755). Many such models have hierarchical aspects (Sheldon, this symposium). Two structural models of personality recently appeared in the American Psychologist: McAdams and Pals’ (2006) “New Big Five” model and Mayer’s (2005) “Systems Framework” models. In addition, empirically-driven models round out such new frameworks, as theorist-researchers add levels and detail. For example, Sheldon, so as to study optimal personality, has employed a six-level model of personality, arranged along a molecular-molar hierarchy (e.g., Sheldon, 2004). More than one good structural model of personality is possible, just as more than one kind of map of a territory is possible: as with political maps, geological maps, and population maps (Mayer, 2001). The first portion of the talk begins with a brief tutorial on structural models, including one set of criteria for their evaluation (Mayer, 2001). Among good models there often is a tradeoff between simplicity-of-understanding and fidelity-to-the-personality-system. With this in mind, two good structural models will be compared: the New Big Five Model (McAdams & Pals, 2006) and the Systems Framework models (Mayer, 2005). The New Big Five Model is focused on five areas of personality, although an alternative interpretation, as a bulls-eye model of personality, is possible as well. The Systems Framework is introduced as a model connecting personality to its surrounding systems. The Framework’s pair-mate model – the Systems Set, organizes inner personality traits. Finally, the merits of the New Big Five and Systems Framework models will be discussed.

Symposia 4: Friday, July 17 11:30 a.m.-12:45a.m. Heritage Ballroom

4. Personality Change and Psychopathology: Effects of Onset, Persistence, and Environmental Interventions

Brian M. Hicks, and Ana C. DiRago, University of Minnesota
Wendy Johnson, University of Edinburgh

Daniel M. Blonigen, Department of Veterans Affairs Palo Alto Health Care System and Standford University School of Medicine

Unraveling the link between personality and psychopathology is a challenging, but important task to enhance our understanding of both personality processes and the etiology of mental illness. Several models have been proposed to account for the personality-psychopathology association including: personality is risk factor for a disorder, the “scar” model wherein mental disorder alters personality, and the continuum model in which personality and disorders are different manifestations of a common process. Importantly, both personality traits and mental disorders are developmental constructs that exhibit normative changes, onsets, and courses. In four longitudinal studies, we examine the impact of psychopathology on normative personality development with a focus on the transition from adolescence to young adulthood. DiRago and Hicks examine the impact of an early (vs. later) onset and persistent (vs. desistent) course of major depression and alcohol dependence on normative personality change. Johnson examines the environmentally mediated impact (i.e., controlling for genetic factors) of adolescent antisocial behavior and substance abuse on personality in young adulthood. Finally, Blonigen demonstrates the effectiveness of treatment interventions for substance abuse to contribute to long-term personality change. Across studies a consistent pattern of convergent and discriminate associations emerged such that personality traits differ in their sensitivity to the impact of psychopathology and environmental experience on normative change. Findings also illustrate the dynamic nature of personality over the life course and the importance of examining normative developmental processes to the understand psychopathology.
Note: No formal discussant is planned. Brian Hicks will provide introductory remarks (about 5 minutes) on the topic that will provide a general overview on normative personality change and prevalence of disorder of interest and then present questions the talks will address. This will minimize redundancy of the introductory material of each talk so that presentations can be more content and data focused. Talks are anticipated to take 15 minutes with time for questions. Time permitting, Brian Hicks will then take a a few (3-5) minutes to make a few brief summary and integrative remarks regarding the talks with any remaining time left for questions.

4.1. Depression and Personality Change during the Transition from Adolescence to Adulthood

Ana C. DiRago, Brian M. Hicks, William G. Iacono, and Matt McGue,University of Minnesota

The prevalence of Major Depressive Disorder (MDD) increases dramatically during the transition from adolescence to young-adulthood. Certain personality traits are associated with increased risk for developing MDD. Specifically, there are strong cross-sectional and prospective associations between high negative emotionality/neuroticism and MDD. We explored how MDD affects personality development using the context of normative change as a frame of reference. To address the interplay between personality and MDD in emerging adulthood, we utilized a large sample of male and female twins participating in the longitudinal Minnesota Twin Family Study (MTFS) assessed from age 17 to 24. Personality was assessed using the Multidimensional Personality Questionnaire data (MPQ) which includes 3 higher-order factors: Positive Emotionality (PEM), Negative Emotionality (NEM), and Behavioral Constraint (CON). Participants were classified into groups based on onset (early vs. later) and course (acute vs. chronic) of MDD. NEM at age 17 was associated with both an early and later onset of MDD. An early and later onset of MDD did not affect normative declines in NEM from age 17 to 24. However, women with an early onset and chronic course of MDD failed to decline on NEM from age 17 to 24. PEM did not act as a vulnerability factor, but instead showed state effects in relation to MDD. Low CON at age 17 was associated with an early onset, but not later onset, of MDD. Additionally, an early onset and acute course of MDD was associated with greater increases in CON from age 17 to 24 relative to other MDD groups. Results illustrate the differential associations between MDD and different personality processes, in particular, the joint effects of an early onset and chronic course of MDD.

4.2. Effects of Onset and Persistence of Alcohol Use Disorder on Personality Development in Young Adulthood

Brian M. Hicks, William G. Iacono, and Matt McGue, University of Minnesota

Several studies have documented the link between alcohol use disorder (AUD) and the personality traits of Negative Emotionality (NEM) and lack of Constraint (CON). Few prospective studies, however, have investigated the impact of AUD on normative personality development, in particular, the unique effects of an early (vs. later) onset and a persistent (vs. desistent) course of AUD. In a sample of 530 men assessed from age 17 to 29, we sought to delineate the link between personality and AUD, specifically, the effects of an early onset (AUD present at age 17) and a persistent course (continued AUD to age 29). NEM and low CON at age 17 predicted both an early and later onset of AUD. Also, low CON at age 17 predicted a persistent course of AUD, and a persistent course of AUD was associated with lower CON throughout young adulthood. Additionally, men with an early onset and persistent course of AUD failed to experience a normative decline in NEM from age 17 to 29. At age 29, the personality of men with a desistent course of AUD was indistinguishable from that of men who never experienced an AUD. The findings suggest that personality traits represent important etiological processes in AUD. Also, while early onset and persistent AUD can alter normative personality development, individuals can recover to attain normative personality functioning if they are able to desist from AUD.

4.3. How Does Behavior Contribute to Personality Development? Using Twins to Identify Mechanisms

Wendy Johnson, University of Edinburgh

The transition from adolescence to young adulthood is characterized by a search for constructive activity that generates financial independence and personal engagement and a social network that generates emotional and sexual satisfaction, which can be summarized as a place in the world. It is also characterized by experimentation with substance use and antisocial behaviors. Personality contributes to involvement in these activities, but its development is also shaped by involvement in these activities. Well-being and (lack of) Alienation are two personality traits that serve as markers of successful progress through this life period. Using Minnesota Twin Family Study data on female twins between the ages of 17 and 24, I explore how genetic and environmental influences on changes in both illicit and antisocial behaviors and attitudes related to the search for a place in the world transact with the development of these important personality traits. I will focus on distinguishing the environmental effects of behavioral choices (alcohol use, antisocial behavior) and life attitudes (boredom, leadership, and positive activity engagement) from intrinsic individual characteristics that contribute to those behavioral choices and attitudes in the first place.

4.4. Treatment, Alcoholics Anonymous, and 16-Year Changes in Impulsivity and Legal Problems among Men and Women with Alcohol Use Disorder

Daniel M. Blonigen, Christine Timko, Bernice S. Moos, and Rudolf H. Moos, Department of Veterans Affairs Palo Alto Health Care System and Standford University School of Medicine

The link between impulsive personality traits and alcohol use disorder (AUD) is well established. However, no studies have investigated whether receipt of help to reduce AUD symptoms (participation in Alcoholics Anonymous [AA] or professional treatment) predicts change in impulsivity and whether such change is associated with relevant outcomes such as legal problems. The present study examined prospective associations between duration of help for AUD and impulsivity and legal problems over 16 years in men and women with AUD. Participants initially untreated for their AUD (NMen = 332, NWomen = 296) completed follow-up phone interviews at 1- and 16-years after their baseline assessment. Impulsivity and legal problems declined between baseline and the 1-year and 16-year follow-ups. After controlling for demographic variables, impulsivity, and legal problems at baseline, a longer duration of participation in AA predicted a decline in impulsivity at both follow-up assessments. In turn, a decline in impulsivity predicted a decline in legal problems at Years 1 and 16. In addition, a longer duration of participation in AA predicted fewer legal problems at Year 1, and this association was moderated by gender (significant in men) and impulsivity (significant for individuals with higher baseline scores). Results highlight the effectiveness of interventions to reduce AUD symptoms as well as concomitant outcomes of behavioral disinhibition (impulsivity, legal problems), and suggest the necessity of such change for prolonged remission of AUD.

Symposia 5: Friday, July 17 2:00 p.m.-3:15 p.m. Grand Ballroom

5. Investigating the Personality of President Barack Obama

Ed de St. Aubin, Symposium Chair, Marquette University
Leslie Hauser and Jefferson Singer, Connecticut College
James Anderson, Northwestern University
Alan Elms, Symposium Discussant, University of California – Davis

This proposal directly responds to the call for symposia that address the connections that personality psychology has to other intellectual traditions. All three papers focus on the personality of President Barack Obama. Each is an application of personality psychology to political psychology and each demonstrates the value of integrating biographical studies with personality psychology. The timeliness and importance of this topic are immense. In a clear demonstration of the vitality and relevance of personality psychology within both intellectual and public domains, these papers begin to help us understand the values, motives, and self-image of the man who will greatly shape the course of our country. As we see in the psychobiographical work of Erikson (1958; 1969) there is a bi-directional dynamic between idiographic and nomothetic modes of inquiry. Applying personality theory to the single case greatly illuminates that life but this intense inquiry should also inform our understanding of personality variables - new insights transformed into hypotheses that might be tested nomothetically. In the first paper, Leslie Hauser and Jefferson Singer quantified the magnitude of three motives as these exist is campaign speeches of Barack Obama and John McCain. Key differences between the two candidates exist. Further, Obama’s motive profile is quite similar to two former presidents, suggesting some possible parallels. Next, Ed de St. Aubin will provide a comparative analysis of Barack Obama and George W. Bush in terms of how core personal values are manifested in one’s identity and how this shapes the presidential self. In the third paper, James Anderson traces the shifting ego-ideal dynamics over Obama’s life span. Anderson suggests that much of this was based on Obama’s perceptions of his father, which changed dramatically while he was an emerging adult. Finally, we are fortunate to have secured Alan Elms as the Discussant. He will provide integrative commentary and then we will have a ten minute Q&A.

5.1. Comparing Barack Obama’s and John McCain’s Motive Profiles from Campaign Speeches

Leslie Hauser and Jefferson Singer, Connecticut College

This paper applies David Winter’s motive profile scoring system to the campaign speeches of Barack Obama and John McCain in order to track how their campaigns evolved and to generate predictions for how Obama might conduct his early presidency. Using coding systems for Achievement, Affiliation, and Power, speeches at three points in their campaign were scrutinized – their early stump speeches (December 2007 and February 2008), their acceptance speeches (August and September, 2008) and their post-convention stump speeches (October, 2008). Two raters were trained in the Winter scoring system and achieved over 90% agreement; a third trained rater also scored a portion of the speeches and achieved a similar level of agreement. Raw motive scores per 1000 words were converted to standardized scores for the two candidates. Across all three speeches, Obama displayed higher Achievement and lower Power than McCain. McCain’s Power increased over time, while Obama’s Power declined. In a critical contrast, McCain’s Affiliation started at a higher level than Obama’s, but declined sharply in his October speech to a level far below Obama’s. Obama’s Affiliation rose in his Acceptance speech and stayed high as the campaign ended. These findings suggest a crucial shortcoming in the McCain strategy as he drifted into an aggressive and more combative rhetoric that contrasted with Obama’s themes of Achievement and Affiliative unity. With regard to predictions about the Obama presidency, prior presidents with motives highest in achievement have run into problems when they attempted to control and micro-manage the immense challenges of their office. Former presidents Carter and Clinton were both strong Achievement types and needed to recognize the limits of their goal orientations. It remains to be seen if this same pitfall will plague Obama or whether he will learn to temper his ideals and draw on more Affiliative and Power oriented strategies.

5.2. Barack Obama and George W. Bush through the Lens of Personality Psychology: Ideology, Identity, and the Presidential Self

Ed de St. Aubin, Marquette University

This paper articulates how the different core values of G.W. Bush and B. Obama are manifested in the identity of each man. Further, we examine how this link between values and identity influence the presidential self, a high profile and closely scrutinized public-leader image that is observed in daily interactions with the press, the public, policy makers, and other international leaders. Individual differences in values are best captured by investigating one’s personal ideology, or value-based world view. This most directly informs one’s perspective in life domains wherein beliefs are more salient than facts: Religious views, political orientation, attitudes towards conservationism, assumptions about the nature of the human species, etc. A rich research literature has supported and extended the work in this area of Silvan Tomkins (1964; 1978; 1987) who suggested that individual differences in personal ideology are best captured according to two orthogonal dimensions. Employing this model of ideology, de St. Aubin et. al. (2006) documented links between ideology and identity by demonstrating that different ideology types tell specific kinds of self-defining life stories. This paper first extends that effort by focusing on the link between personal ideology and self identity for the current and former presidents. For instance, the normative (a Tomkins dimension) value system adhered to by Bush is manifested in an identity that emphasizes emotional control, a maintenance of a hierarchy of human worthiness, and a view of the self as unchanging. The humanist ideology of Obama has is seen in an identity that is contoured by interpersonal connectedness (note the findings in paper #1 above regarding Obama’s affiliative motive) and a tendency for intentional selfing that views the self as a complex, organic being to be explored and nurtured. We then turn to the presidential self and address the ways in which each man portrays his public-leader image – including support for or against specific policy - in a way that highlights this congruence between values and self.

5.3. Barack Obama’s Image of a Leader and His Relationship with His Father

James William Anderson, Northwestern University

Barack Obama has an image of how he would like to be, his ego ideal or idealized self-image. I will explore how his shifting images of his father, Barack Obama Sr, played a central role in the development of his ego ideal. Obama formed an image of his father based largely on stories he heard about him. He pictured his father as being fair, strong, charismatic, and fearlessly believing in what was right, and he internalized this image as a fundamental aspect of his ego ideal. “It was into my father’s image,” Obama (1995, p. 119) recalled, “…that I’d packed all the attributes I sought in myself.” When he saw other men fail to meet up to the “lofty standards” of this image, his father’s voice within him “remained untainted, inspiring, rebuking, granting or withholding approval.” In the period after he graduated from college, he learned that his father was sharply different from the image he had had of him. He understood that his father in Kenya had been a failure, in particular because he was inflexible, could not get along with other people, and acted harshly and arrogantly toward others. Obama internalized an image of a strong, fair person with deeply held values, an image largely based on how he imagined his father was. By the time he learned that his father had not lived up to that image, the image had become part of his ego ideal. But then he supplemented his image with a further expectation of how he would behave. This added part of his ego ideal was based on his father as a counterexample; in other words, Obama would be pragmatic and would seek to get along with people, as his father had not. This unusual experience, of combining the image of the idealized father with the image of a father’s whose failings he later learned of, helps explain the kind of leader Obama has become.

Symposia 6: Friday, July 17 2:00 p.m.-3:15 p.m. Heritage Ballroom

6. Personality and Economics

Ulrich Schimmack, University of Toronto Mississauga Canada

Richard E. Lucas, Michigan State University

Standard economics has relied on a simplistic model of human behavior (homo economicus). Personality psychology has focused on internal dispositions as explanations of human behavior. Not surprisingly, economics and personality psychology have developed independently. However, the world is changing. Economists are recognizing that humans are more complex and differ from each other and personality psychologists have started to examine how personality dispositions influence real-world outcomes. One of the most exciting new developments is the collaboration of personality psychologists and economists in longitudinal (panel) studies. In these studies, large national representative samples of households answer survey questions once a year. This symposium features a series of papers that present research from the German Socio-Economic Panel (SOEP) study of the German Institute of Economic Research (DIW). In 2005, the SOEP included a brief Big Five measure. These data provide an unprecedented opportunity for collaboration between personality psychologists, economists, and other social scientists. The first paper shows how longitudinal panel data can be used to test the hypothesis that self-ratings of the Big Five personality traits actually measure stable dispositions (Ulrich Schimmack). The second paper examines how personality of spouses influences well-being (Richard E. Lucas). The third paper examines the relation between income and life-satisfaction using mixture distribution latent state-trait modeling (Michael Eid). The fourth paper shows how household panel data can be used to examine genetic and environmental influences by enriching panel data by oversampling twins (Frank Spinath).The SOEP panel data and similar data sets from the UK and Australia can be analyzed by international personality psychologists. This novel opportunity to examine old and new questions in personality research using rich, longitudinal data, with large national representative samples is likely to have a profound influence on the future of the field and is going to create exiting opportunities for collaboration between personality psychologists and economists.

6.1. A Trait-State-Error Model of Life-Satisfaction and the Big Five Personality Traits

Ulrich Schimmack, University of Toronto Mississauga, Canada

The existence of stable personality dispositions is a core assumption of personality psychology. Typically, stable personality traits are measured with self-report measures to report their dispositions. Although these questionnaires show high stability over time, it is questionable that people are perfect judges of these stable dispositions. The trait-state-error model (TSE, Kenny & Zautra, 1995) provides an alternative approach to the assessment of stable dispositions. The model uses the pattern of retest correlations over a minimum of four occasions to test whether the variable under investigation is influenced by a stable disposition and estimates its contribution to the reliable variance on a specific occasion. I used the TSE model to demonstrate that approximately 50% of the reliable variance in life-satisfaction is explained by a stable disposition (Schimmack, Krause, Wagner, & Schupp, in press). In this paper I examine the relation between the trait, state, and error variance in life-satisfaction and a measure of the Big Five that was assessed on one occasion in 2005. Consistent with the interpretation of this measure as a measure of personality traits, the model shows associations between Big Five dimensions and the trait-component of life-satisfaction. However, the model also shows that the Big Five dimensions are related to state variance in life-satisfaction. This finding shows that the self-report measure of the Big Five is not a pure measure of stable dispositions. Associations with the error component were weak, replicating other findings that occasion-specific measurement error has negligible effects on personality ratings (Eid & Diener, 2004). In short, the paper shows how the SOEP data can be used to test a core assumption of personality psychology that the Big Five measure stable personality traits.

6.2. Predicting Life Satisfaction from Relationship Partners’ Personality: The Importance of Actor, Partner, and Similarity Effects

Richard E. Lucas, Michigan State University

Portia S. Dyrenforth, Hobart and William Smith Colleges

Deborah A. Kashy, and M. Brent Donnellan, Michigan State University

Why are some people satisfied in their marriages, whereas others are dissatisfied? Do the personality traits of the relationship partners play a role in their relationship and life satisfaction? Although numerous studies have attempted to answer these questions, contradictory results have often emerged. However, existing studies tend to use relatively small samples of couples to answer these questions. Thus, many existing studies may not have enough power to detect subtle effects of personality and personality similarity. Furthermore, couples who volunteer for studies on relationships might differ in important ways from a more representative sample. Together, these factors may contribute to inconsistency in the literature. In this study, we use data from a representative sample of 5,709 couples who participated in the German Socio-Economic Panel Study (GSOEP) to examine the relative importance of three types of personality effects on life satisfaction: actor effects (which reflect the association between a person’s personality and his or her own life satisfaction), partner effects (which reflect the association between a person’s personality and his or her spouse’s life satisfaction), and similarity effects. Results indicated robust support for actor effects such that an individual’s personality is associated with his or her life satisfaction. Partner effects also emerged showing that the personality traits of a spouse are associated with life satisfaction. Small similarity effects emerged when similarity indices were entered alone in a regression equation predicting satisfaction. However, similarity explained very little variability when the more appropriate technique of first controlling for actor and partner effects was used. Thus, although husbands’ and wives’ personalities are related to each other’s life satisfaction, the similarity of their personalities is not.

6.3. Income and Life-Satisfaction

Maike Luhmann and Michael Eid, Free University Berlin, Germany

The relation between income and happiness is an old question. Numerous articles claim that money does not buy happiness. Some articles suggest that dispositional happiness is a predictor of higher income in the future. However, most of the evidence is limited to cross-sectional studies or longitudinal studies with two occasions. This paper examines the relation between income and life-satisfaction in the SOEP over 17 years of annual assessments. The data are analyzed with a mixture distribution latent state-trait model. The main findings are that (a) both life-satisfaction and income are influenced by stable dispositions (trait variance) and that the correlation between these two dispositions contributes to the correlation between income and life-satisfaction. The model also shows that changes in income are associated with changes in life-satisfaction (state variance). This finding suggests that becoming richer is associated with becoming happier. The paper also examines similarities and differences between East and West Germany to examine how much parameter estimates are affected by historic events like the transformation of East Germany from a socialist state into a Western market economy and democracy.

6.4. Personality Psychology and Panel Studies: Improvements and Future Challenges in the Field of Genetically Sensitive Sample Designs

Frank M. Spinath, Saarland University, Germany

Understanding the sources of individual differences beyond social and economic effects has become a research area of growing interest in personality psychology, sociology, and economics. A quantitative genetic research design provides the necessary tools for this type of analysis. For a state-of-the-art approach, multigroup data is required. Household panel studies, such as BHPS (Understanding Society) in the UK or the SOEP in Germany, combined with an oversampling of twins, provide a powerful starting point since data from a reasonably large number of non-twin relatives is readily available. In addition to advances in our understanding of genetic and environmental influences on key variables in the social sciences, quantitative genetic analyses of target variables can guide molecular genetic research in the field of personality, employment, earnings, health and satisfaction, as combined twin and sibling or parent data can help overcome serious caveats in molecular genetic research. The presentation highlights future advances and presents preliminary data from combined twin and household samples.

Symposia 7: Friday, July 17 3:30 p.m.-4:45 p.m. Grand Ballroom

7. Digging Deeper: Probing the Connection Between Personality and Physical Health

Co-Chairs: Grant W. Edmonds , University of Illinois and Daniel K. Mroczek, Purdue University
Presenters:
Grant Edmonds , University of Illinois, Urbana-Champaign
Timothy W. Smith, University of Utah
Sarah E. Hampson, Oregon Research Institute, Eugene

Daniel K. Mroczek, Purdue University
Discussant: Robert Wilson , Rush University Medical School

The connection between personality and physical health is well established. Recent research has sought to build on this basic finding by probing the different ways in which personality predicts physical health. The four papers that make up this symposium are similar in that all use physical health as an outcome (coronary atherosclerosis, mortality, SF-36 and LISRES ratings, and global clinical health status). However, each uses a different type of personality predictor, thus illustrating the myriad and complex ways that personality influences health. Edmonds, Jackson & Roberts show how changes in conscientiousness predict changes in physical health as assessed by the SF-36 and LISRES. Smith documents that spousal rating of personality do a better job of predicting coronary atherosclerosis (coronary calcification) than self-ratings. Hampson, Goldberg, Dubanoski & Hillier demonstrate how childhood ratings of personality traits predict clinically-assessed global health through cognitive ability over a 50 year longitudinal period. Finally, Mroczek, Turiano & Spiro illustrate how neuroticism is mediated by health behaviors in predicting mortality over a 30-year longitudinal follow-up. Each paper shows that the way we construe our personality constructs makes a great deal of difference when predicting physical health. Sometimes spouse ratings work better (Smith). Sometimes the effect of personality is sharpened and clarified by mediators such as health behaviors or cognitive ability (Hampson, Mroczek). Sometimes changes in personality exert important effects on health or changes in health (Edmonds). Robert Wilson, a medical school-based neuroscientist, will serve as discussant providing unique insights from the perspective of clinical bench science. Lastly, these four papers, as well as the discussant, exemplify the theme of this year’s ARP conference: the rich connections between personality and other disciplines.

7.1. Prospective Associations Between Childhood Personality Traits and Adult Health Outcomes in the Hawaii Personality and Health Cohort

Presenter: Sarah E. Hampson , Oregon Research Institute, Eugene, Oregon and University of Surrey, Guildford, UK

Presenter: Lewis R. Goldberg, Oregon Research Institute, Eugene, Oregon

Joan P. Dubanoski, Kaiser Permanente Center for Health Research, Hawaii

Teresa A. Hillier, Kaiser Permanente Center for Health Research, Hawaii and Northwest

Childhood Big Five personality traits, particularly Conscientiousness, predict longevity and one proposed pathway for these relations is through morbidity. Cognitive ability is another robust predictor of longevity whose influence may be mediated by morbidity. We are studying relations between childhood traits and midlife cognitive ability and health status in the Hawaii Personality and Health Cohort, which is a community sample of over 2,000 individuals from two Hawaiian islands who underwent a personality assessment by their elementary school teachers 40-50 years ago. To date, 83% of the original sample has been located and 70% of them have participated in follow-up studies. Childhood personality traits predicted adult self-reported health and health behaviors in analyses of 1,057 participants. A subset of these (n = 555) have also completed an extensive medical and psychological assessment at mean age 50. So far, in this subset, childhood traits do not predict an index of global clinical health status based on biomarkers assessed at the clinic examination. However, childhood Extraversion, Conscientiousness, Emotional Stability, and Openness for men, and childhood Conscientiousness and Openness for women, predict one or more of three aspects of cognitive ability measured at age 50 (verbal comprehension, concept formation, and visual matching). These childhood traits may exert an influence on longevity indirectly through their influence on the development of cognitive abilities. Consistent with this model, higher Concept Formation ability was associated concurrently with better clinically assessed global health, and higher Verbal Comprehension was associated with better self-rated health (Radj = .19 - .23). These and other findings from this cohort will be used to illustrate some of the complexities of tracing the influence of childhood traits on life-course pathways to health and longevity.

7.2. Do Health Behaviors Explain the Effect of Neuroticism on Mortality?

Daniel K. Mroczek, Purdue University, West Lafayette, IN

Nicolas Turiano1 & Avron Spiro, Boston VA Healthcare System, Boston, MA

Studies have shown that higher levels of neuroticism are associated with earlier mortality. Yet what accounts for this association? One explanation is that persons higher in neuroticism engage in certain poor health behaviors, specifically smoking and excessive drinking, thus leading to earlier death. We tested this hypothesis using 30-year mortality in 1,788 men, from the VA Normative Aging Study. Using proportional hazards (Cox) models, we found that including smoking as a predictor attenuated the effect of neuroticism on mortality by 40%. The effect of drinking was not clear, and it appeared to act as a suppressor variable, raising the effect of smoking but not having an effect on mortality in and of itself. Considering that 60% of the neuroticism-mortality association remained unexplained, this suggests that the effects of other pathways (e.g, biological) also influence the relationship. This study provides evidence for pathways leading from the personality trait of neuroticism to mortality via specific health behaviors. At the same time it highlights the likelihood that health behaviors are not likely to be the only modality connecting personality to mortality.

7.3. Changes in Conscientiousness Predict Changes in Physical Health

Grant W. Edmonds, Joshua Jackson, and Brent W. Roberts, University of Illinois, Urbana-Champaign

The personality trait of conscientiousness has well-known relations to positive health, with higher scores being related to better health behaviors and better physical health. Conscientiousness also shows conspicuous increases across the life course including old age, which invites the question of whether changes in conscientiousness predict gains in health above and beyond original standing on conscientiousness. We tested this idea in a 3 year longitudinal study of Illinois residents (N = 317). Using path analysis, we found that changes in conscientiousness predicted experiencing fewer health problems in the 3 years of the longitudinal study (B = .21, p < .05), above and beyond original standing at time 1, while controlling for age, income, and obesity. We found similar patterns with overall health and physical limitations. I will discuss the implications these findings have for positive aging, health, and personality change in old age.

7.4. Where should we look and who should we ask when studying personality and health?

Timothy W. Smith, Department of Psychology, University of Utah

Personality traits predict important physical health outcomes, demonstrating the utility of personality constructs and measures. Better use of personality science can help refine our understanding of psychosocial influences on health. Often individual measures are used without concern about overlap with the many traits and scales studied elsewhere, thereby impeding a more systematic and cumulative science of personality and health. Further, reliance on self-reports of personality raises concerns about mono-method biases. For example, aspects of negative affect (i.e., anxiety, depressive symptoms, anger) and social behavior (i.e., hostility, dominance) are coronary risk factors, but typically these traits are examined individually and through self-reports, without consideration of their potentially overlapping associations with disease or participants’ willingness and ability to provide accurate assessments of undesirable traits. This presentation uses a recent study of asymptomatic coronary artery disease (CAD) (i.e., coronary calcification assessed non-invasively via CT) in 154 otherwise healthy married couples (mean age = 64) to illustrate the value of established personality frameworks and multiple methods. Participants completed NEO-PI-R self-reports and spouse ratings of negative affect (e.g., anxiety, depression, anger), dominance, and affiliation, the latter two scales from an interpersonal circumplex (IPC) re-scoring of the NEO. Each affective trait was significantly associated with CAD when considered separately, but only anxiety and anger were significant in simultaneous tests. Affiliation was inversely related to CAD, and dominance was positively associated. Effects of (low) affiliation and anger were overlapping, whereas effects of anxiety and dominance were independent. Importantly, effects were significant for spouse-ratings, but generally not for self-reports. Established structural models of personality (i.e., FFM and IPC) can guide a more systematic and cumulative understanding of personality and health, and methods beyond self-report may illustrate further limitations in the present literature. Reliance on self-reports of personality might contribute to an under-estimate of associations between personality and health.

Symposia 8: Friday, July 17 3:30 p.m.-4:45 p.m. Heritage Ballroom

8. Advances in Research on Personality and Social Relationships

Chair: Lars Penke, The University of Edinburgh
Speaker: Jaap Denissen, Humboldt University of Berlin

Peer-Rated Big Five Reputations as Longitudinal Predictors of Group
Outcomes: A Functional Analysis
Speaker: Mitja Back, University of Leipzig

From First Sight to Friendship: The Role of Initial Attraction and Personality

Speaker: Lars Penke, The University of Edinburgh
Effects of Long-Term and Short-Term Relationship Interests on Actual Mate Choices in the Berlin Speed Dating Study
Speaker: Michela Schröder-Abé, Chemnitz University of Technology
Make You Feel My Love: Emotional Intelligence and Relationship Quality

The interplay between personality and social relationships is as interesting as it is complex: On the one hand, real-life social phenomena can usually not be understood without taking the personality of the involved individuals into account. On the other hand, many personality traits can only be expressed and developed in social settings. This symposium presents four different research projects from four members of the scientific network "Personality and Social Relationships: State of the Art and Perspectives", which is funded by the German Research Foundation (DFG). While they study the effects of personality on very different real-life social processes (group formation, friendship development, mate choice, and relationship maintenance, respectively), they share a strong focus on advanced statistical methods for the analysis of personality effects on social relationships, which are based on David Kenny's Social Relationships Model (SoReMo).

8.1. Peer-Rated Big Five Reputations as Longitudinal Predictors of Group Outcomes: A Functional Analysis

Jaap J. A. Denissen, Humboldt University of Berlin, Germany

Maarten H. W. Selfhout, Utrecht University, the Netherlands

The vast majority of empirical research on personality has focused on the so-called actor perspective on personality, operationalized by people's self-ratings. In contrast, much less is known about the so-called observer perspective, operationalized as people's trait reputations within social groups. From a socioanalytic perspective (Hogan, 1996), traits are described as evaluative categories that observers use in order to describe an individual's reputation. Such reputations are thought to have an evolutionary function because they can forecast an individual's usefulness in the economy of its social group. In an empirical study, this theoretical framework was investigated by collecting Big Five personality trait ratings within a longitudinal round-robin design of 225 individuals, who first met each other and then interacted regularly within the context of work groups of 20-25 members. In addition, ratings of other people's domain-specific (social, emotional, instrumental) and general utility value were collected. As dependent variables, ratings of friendship and group influence were assessed. Findings supported the predictive validity of utility ratings, as it was actually beneficial for group members to associate with highly valuable group members over time. In addition, Big Five reputations were meaningfully related to indicators of domain-specific and general utility value, which were in turn associated with horizontal (friendship) and vertical (leadership) group outcomes. These findings have important implications for conceptualizations of traits from a functionalist, evolutionary perspective. Moreover, they are consistent with a socioanalytic perspective by showing that traits have both social antecedents and consequences and should not only be investigated using self-reports.

8.2. From First Sight to Friendship: The Role of Initial Attraction and Personality

Mitja D. Back, Department of Psychology, University of Leipzig, Germany

Stefan C. Schmukle, Department of Psychology, Westfälische Wilhelms-University Münster, Germany

Boris Egloff, Department of Psychology, University of Leipzig, Germany

Since the days of the ancient Greek philosophers, friendship has been conceived of as one of the most relevant forms of social relations. However, what leads to friendship is still not very well understood. We analyzed the role of initial attraction as well as personality traits in predicting friendship. Interpersonal perceptions like friendship may be decomposed into a perceiver effect (harshness of each perceiver), a target effect (popularity of each target) and a relationship effect (unique relational attraction to a person). Accordingly, first impressions and personality may influence friendship with respect to these three components. We investigated a group of psychology freshmen upon encountering one another for the first time. Personality traits and attraction ratings were assessed using a large round-robin design (2,550 dyadic ratings). One year later, the same students rated each other concerning their mutual relationships. Results show that initial attraction has a meaningful long-term influence on friendship development on the level of the perceiver, the target and the relationship. Moreover, personality was an important determinant for each component of friendship: It predicted who is friend with others (perceiver), who is seen as a friend (target) and who is friend with whom (relationship).

8.3. Effects of Long-Term and Short-Term Relationship Interests on Actual Mate Choices in the Berlin Speed Dating Study

Lars Penke, The University of Edinburgh, United Kingdom

Mitja D. Back, Department of Psychology, University of Leipzig, Germany

Jens B. Asendorpf, Humboldt University of Berlin, Germany

Research on romantic attraction and mate choice usually relies on self-reports, reactions to isolated cues, or observations in artificial laboratory situations. In contrast, speed dating offers a unique environment to systematically observe initial romantic attraction, mate choice, and courtship in real life. In the Berlin Speed Dating Study, a community sample of 382 singles with broad age range participated in speed dating events under experimentally controlled conditions. Individual characteristics were assessed and two follow-up studies were conducted after 6 weeks and 1 year. Here we will present multilevel Social Relations Model (SoReMo) analyses, which are able to separate actor, partner and relationship-specific effects, on the role of long-term vs. short-term relationship interests in actual mate choices and their long-term consequences. Results indicate that being generally more popular at speed datings (partner effects) was mainly predicted by various attractiveness components, though status indicators had incremental effects in men. Both general long-term and short-term interests, which reflected the personality traits of sociosexuality and shyness, influenced general choosiness (actor effects) in men and women, as well as relationship outcomes in the following year. However, effects on choosiness were moderated by self-perceived mate value in men and disappeared in women who are in the fertile phase of their menstrual cycles. Instead, choices of fertile women appeared to become more specific (i.e., driven by relationship effects) and guided by similarity on certain characteristics as well as interactions between female sociosexuality and male attractiveness.

8.4. Make You Feel My Love: Emotional Intelligence and Relationship Quality

Michela Schröder-Abé, Andrea Bräunig, & Astrid Schütz
Chemnitz University of Technology, Chemnitz, Germany

Meyer and Salovey (1997) identified the ability to recognise, understand, use, and regulate emotions as four interrelated aspects of emotional intelligence (EI). Although theorists have put forward that EI should be linked to the perceived quality of relationships, little empirical research has systematically examined EI in romantic relationships using appropriate dyadic designs and analyses. Furthermore, recent studies failed to find partner effects of EI on relationship satisfaction and love, despite conceptualizations of EI suggesting that a person's EI should influence his/her partner's reactions. The present study investigated the relationship between EI and relationship quality (RQ), expecting positive actor as well as partner effects of EI on RQ. Thereby relationship satisfaction, trust, love, commitment, and emotional closeness were measured as independent aspects of RQ, as past research has shown that perceived RQ components are domain-specific constructs. EI was measured with the Wong and Law EI Scale. This self-report scale assesses four emotional abilities: appraisal and expression of one's emotions (AS), appraisal and recognition of emotion in others (AO), use of emotion to facilitate performance (UE), and regulation of emotion in oneself (RE). 191 heterosexual couples (12 % married, 56 % students, mean relationship duration 60 months) participated in the study via internet. The results confirmed our hypotheses: In addition to actor effects, we also found partner effects of EI predicting RQ. For AS, AO, and UE, significant actor effects were found predicting all aspects of RQ, whereas RE only produced significant actor effects in men. The most pronounced partner effects were found for AO, an interpersonal emotional ability. Emotional closeness was the aspect of RQ most closely connected to EI. The results are among the first to show effects of EI in self- and especially partner perceptions of RQ and attest to the important contribution of EI to optimal social functioning.

Poster Session #1. Friday, July 17, 4:45-6:00 PM. Grand Ballroom

1. Critically Evaluating the Narcissistic Personality Inventory: Conceptual and Psychometric Considerations

Robert A. Ackerman, Edward A. Witt, and M. Brent Donnellan, Michigan State University

The Narcissistic Personality Inventory (NPI; Raskin & Terry, 1988) is perhaps the most widely used measure of normal narcissism in social and personality psychology. However, there are emerging and existing concerns with the response format, factor structure, content validity, and construct validity of this measure. We evaluated these issues in series of three studies. Studies 1 and 2 examined issues related to the forced-choice nature of the NPI. An implicit assumption guiding the NPI questions is that the two options reflect opposite ends of the same continuum. We evaluated this assumption by taking 16 pairs of statements from the NPI and presenting them as individual questions using a 5-point Likert-type response scale (i.e., 32 items). We selected these pairs of statements because they constitute the item pool for the NPI-16 (Ames, Rose, and Anderson, 2006). Moreover, Ames et al. (2006) selected these items because they represent core features of narcissism. Across the two studies, the mean correlation between item pairs was -.14 (SD = .28). However, we identified several NPI pairs that were positively correlated. We also evaluated whether narcissistic and non-narcissistic composites calculated from these responses hold similar places in a nomological net. In Study 3 we examined the factor structure of the forced-choice NPI and evaluated the convergence of the different NPI facets (e.g., superiority, exploitativeness) with more recently developed measures of normal narcissism and pathological narcissism. All told, this set of studies adds to the literature concerning the assessment of normal narcissism by identifying concerns with the NPI and areas for future scale development.

2. The Up-Regulation of Positive Affect: Cognitive and Behavioral Strategies in Everyday Life

Kimberly M. Angelo and Sanjay Srivastava, University of Oregon

Three studies investigated cognitive and behavioral strategies for up-regulating positive emotions in everyday life, asking: What strategies do people use? Who uses which? What are their short- and long-term affective outcomes? In Study 1, participants (N = 109) listed ways they create or maintain positive emotions. Study 2 (N = 277) examined the structure of these strategies, and examined relations between these strategies and individual differences in personality and well-being. Exploratory factor analysis revealed six broad strategies: social affiliation, direct regulation, attainment focus, imagination, mental stimulation, and comfort activities. Study 3 (N = 148) examined the use of these strategies in daily life using the Day Reconstruction Method (Kahneman et al., 2004), a one-time diary method in which participants divided their previous day into episodes and reported on their use of the six strategies and on their positive and negative affect for each episode. Multi-level modeling revealed that individual differences, particularly extraversion and behavioral activation, predicted overall use of regulation strategies, whereas other traits predicted specific strategies (e.g., openness predicts imagination use). Positive regulation strategies were also associated with both trait and state positive affect. For two strategies, a trade-off emerged between person-level and state-level affect: Comfort activities were associated with short-term positive affect and long-term negative affect, whereas the opposite pattern was found for attainment focus. Across studies, social affiliation and imagination were positively correlated with both state and trait positive affect. These results suggest that up-regulation of positive emotion can be achieved through the use of a variety of behavioral and cognitive strategies, and that these strategies have differential effects on short- and long-term well-being.

3. Personality associations of visual processing style: Preliminary results

Elizabeth J. Austin and Ya-Shyuan Jin, University of Edinburgh, UK

In two studies participants (N = 52, 48) completed mood and personality scales and viewed a series of positively-, negatively- and neutrally-valenced pictures (10 of each type) whilst their eye movements were tracked. Each picture was presented on a patterned background and was displayed for 10s. In Study 1 half the participants were instructed to manage their mood whilst viewing the pictures and half were instructed to view the pictures naturally. As no effect of instruction condition was found, results from both conditions were combined for analysis. In Study 2 all participants were instructed to view the pictures naturally. Visual processing parameters were found to be consistent across stimulus valence. Correlations for mean fixation duration ranged from .66-.92 and for mean saccade amplitude (angular distance between fixations) from .75-.90, indicating the existence of consistent individual differences in visual processing style. There was less consistency for proportion of time spent fixating on the picture as opposed to the background (r range .28-.75). Regression modelling using mood and personality as predictors showed that BAS Reward-Responsiveness (Study 1) and Dysfunctional Impulsivity (Study 2) were negatively related to mean saccade amplitude, and BASRR was also negatively associated with mean fixation duration in Study 1. These results provide preliminary indications that individual differences in visual processing style may be a behavioural marker for BAS-related traits, with high BASRR and Dysfunctional Impulsivity scorers having a viewing style characterised by short fixation durations and short saccades. Further research is required in order to understand the basis of these observations (e.g. to examine how are these associations are moderated by different stimulus types and by the incorporation of cues to reward-seeking or impulsive behaviour in the viewing instructions).

4. Narcissism and social comparison: The role of self-esteem

Dick P.H. Barelds and Pieternel Dijkstra, University of Groningen, The Netherlands

Social comparison is an important, if not central, characteristic of human social life. Especially under conditions of threat or uncertainty, individuals tend to cope by comparing themselves to others in a similar situation. Because individuals with low self-esteem experience inner uncertainty, they are usually more inclined to compare themselves with others. Although narcissism has been related to uncertainty and self-esteem, the relation between narcissism and the inclination to compare oneself remains largely unstudied. On theoretical grounds one might expect narcissism to be related positively to individuals’ tendency to socially compare themselves. Recently, Bogart, Benotsch and Pavlovich (2004) indeed found evidence for a positive relation between narcissism and social comparison tendency in a sample of 109 students (75% female). They also reported that the relation between self-esteem and narcissism was not moderated by social comparison. The correlational pattern in their study, however, suggests that the relations between narcissism, social comparison and self-esteem might be more complex. The present study examined the relations between narcissism, social comparison, and self-esteem in a Dutch community sample of 451 participants (230 men, 221 women; mean age 50 years, SD = 14.7 years, range 20-90). Results show that narcissism was related positively to self-esteem and social comparison, whereas self-esteem and social comparison were related negatively. More specifically, self-esteem was found to act as a suppressor variable in the relation between narcissism and social comparison: controlling for self-esteem increased the relation between narcissism and social comparison. In addition, sex and age differences were found in the relationship between narcissism, self-esteem and social comparison tendency. Theoretical and practical implications of the present study’s findings are discussed.

5. Exploring Morality through Family Dynamics and Individual Differences

Sarah Berger, Cristina Brown, and Brenda McDaniel, Kansas State University

Moral development has been viewed as involving familial, spiritual, societal, and emotional components. The present study examined how these components may work in concert. Sixty participants participated in the present study for course credit. The majority of participants were 18 to 19 years of age and identified themselves as Caucasian. In small groups, participants completed measures of family functioning, empathy, shame, guilt, and ability to forgive. The individual difference variable of empathic concern (i.e., feelings of sympathy and compassion) was shown to be predictive of family health and competence such that higher family dysfunction was related to lower reports of empathic concern [F (1, 58) = 9.91, p = .003, R2 = .15]. Further, shame was shown to be predictive of family cohesion such that lower reports of cohesion were related to higher reports of shame [F (1, 58) = 4.48, p = .039, R2 = .07]. The individual difference variables of shame and ability to forgive were shown to be predictive of personal distress (i.e., distress or discomfort in response to others in distress/emotional flooding/anxiety) such that higher levels of shame and lower levels of ability to forgive were related to higher levels of personal distress [F (2, 57) = 5.88, p = .005, R2 = .17]. On the other hand, guilt was shown to be predictive of empathic concern [F (1, 58) = 9.04, p = .004, R2 = .14]. The present findings shed light on different components of morality. Specifically, it was found that family dynamics impact individual levels of empathy and shame. Furthermore, distinct relationships were found with guilt being related to empathy and shame being related to internal conflict. The present study furthers our understanding of morality, family dynamics, and individual differences.

6. A Self-Regulatory Model of Behavioral Disinhibition in Late Adolescence: Integrating Personality Traits, Externalizing Psychopathology, and Cognitive Capacity

Tim Bogg and Peter R. Finn, Indiana University, Bloomington

Two samples with heterogeneous prevalence of externalizing psychopathology were used to investigate the structure of self-regulatory models of behavioral disinhibition and cognitive capacity. Consistent with expectations, structural equation modeling in the first sample (N = 541) showed a hierarchical model with three lower-order factors of impulsive sensation-seeking, anti-sociality/unconventionality, and lifetime externalizing problem counts, with a behavioral disinhibition superfactor best accounted for the pattern of covariation among six disinhibited personality trait indicators and four externalizing problem indicators. The structure was replicated in a second sample (N = 463) and showed that the behavioral disinhibition superfactor, and not the lower-order impulsive sensation-seeking, anti-sociality /unconventionality, and externalizing problem factors, was associated with lower IQ, reduced short-term memory capacity, and reduced working memory capacity. The results provide a systemic and meaningful integration of major self-regulatory influences during a developmentally important stage of life.

7. Differential Meta-accuracy: People are Aware of the Relative Impressions they Make

Erika N. Carlson, Washington University in St. Louis

R. Michael Furr, Wake Forest University

Simine Vazire, Washington University in St. Louis

While previous research suggests that people are unaware of the different impressions they make (Kenny & DePaulo, 1993), recent research re-evaluating this long-held conclusion found evidence that people do in fact accurately detect differences among others' perceptions of their personality (Carlson & Furr, in press; Psychological Science). In this research, meta-accuracy was assessed for acquaintances from different social contexts of college participants' lives, a design based upon the logic that: a) people tend to behave differently across different contexts, b) interaction partners from different contexts witness these differing behaviors and form differing impressions, and c) meta-perceptions (i.e., beliefs about others' perception of the self) should be differentiated across contexts because contextual information (e.g., others' feedback) is also relatively differentiated. Thus, unlike previous research that focuses on acquaintances from a single context, this contextually-differentiated design captures potentially meaningful differences in the impressions that participants create in people from their real lives. In the current research, we replicate and extend Carlson & Furr's findings by exploring additional traits and potential moderators of meta-accuracy. To assess "differential meta-accuracy" (DMA; degree to which people detect the different impressions they make), participants rated their meta-perceptions for up to three acquaintances who knew them well and those acquaintances then provided their actual views of the participants in terms of the same traits. Replicating Carlson & Furr, results showed that the average person achieved high and significant DMA for each of the Big Five traits. Additionally, the average person was able to detect how he or she was seen on other traits such as intelligent, narcissistic, likeable, and attractive. Thus, when given meaningful variability in impressions to detect, people can indeed detect the relative impressions they make on others suggesting that people have a greater awareness of the variability in their social identity than previously thought.

8. Accuracy of Personality Impression Is Affected By Social Identity and Mode of Information Gathering

Avner Caspi and Sonia Roccas, Open University of Israel

This study aimed at integrating two lines of research that examine social perception: personality perception and group processes. The first line of research emphasizes factors that affect accuracy of perceptions of personality of individuals. The second line of research emphasizes the importance of social identities. Social identities affect both the way a person construes his personality (self stereotyping) and the way others perceive him (stereotyping). 32 students (targets) wrote short open ended self-descriptive paragraphs and completed questionnaires of Personality traits. Participants (raters) were one hundred and sixty students. They read or listened to the open-ended self-descriptions and rated the targets with the same set of questionnaires used by the targets. Each participant rated eight targets. Each target was rated by ten raters. Half of the targets and half of the raters completed the questionnaires after the salience of their identity as psychology students was raised. Findings indicate that self-other agreement was higher for targets' whose social identity was salient when they wrote their self description. This indicates that the salient social identities affect and clarify self presentations. Furthermore, inter-rater consensus was higher when the raters' social identity was salient, indicating that the rater's shared stereotypes affected the perception of the target. Finally, the saliency of the social identity of the rater interacted with the salience of the social identity of the target: consensus was highest when the social identity of both the target and of the raters was salient, and lowest, when the social identity of both target and rater was not salient. These results demonstrated the important influence of social identity salience on social perception.

9. Leading by Fear or Admiration? Personality Predictors of Two Fundamental Leadership Styles

Joey T. Cheng, Jessica L. Tracy, and Joseph Henrich, University of British Columbia

Leadership is the process of exerting influence to attain shared goals (Bass, 1990). Converging lines of research (e.g., Chance and Jolly, 1970; Gilbert, 1989; Henrich & Gil-White, 2001) suggest that there are two broad ways of leading, or attaining the social status necessary to lead: through dominance (i.e., influencing others by force, threat, or intimidation; i.e., fear-based status) or prestige (i.e., influencing others by sharing of wisdom, skills, or expertise; i.e., respect-based status). Previous research has uncovered the personality profiles associated with general leadership tendencies (Anderson et al., 2001). However, it is less clear which personality traits predict the two distinct leadership styles. Based on evolutionary accounts suggesting that the two styles should be associated with divergent interpersonal behaviors (i.e., aggression vs. affiliation), we tested whether personality traits associated with these broad dimensions differentiate between the predisposition to lead through dominance vs. prestige. Small zero-acquaintance groups (4-6 members; N=192) engaged in a 20-minute interactive task. They then rated all group members on dominance, prestige, interpersonal traits (agency and communion), and influence over the group, in a round-robin design. These peer ratings were parsed into actor, perceiver, and dyadic components using the Social Relations Model (Kenny, 1994). Participants also self-reported on self-esteem, narcissism, and academic achievement (GPA). Results demonstrated that: (a) dominance and prestige are fairly independent paths toward high status; (b) self-esteem is positively related to prestige, whereas narcissism is positively related to dominance; and (c) both leadership styles are associated with high agency, but dominance is low communal, and prestige high communal. These findings suggest that personality traits may influence the adoption of leadership strategies, such that individuals (unconsciously) choose dominance or prestige on the basis of which behavioral strategy provides the best fit for their underlying dispositions.

10. Interpersonal perceptions across the social network

Allan Clifton, Vassar College

Models of interpersonal perception (e.g., Kenny, 2004) suggest that consensus in personality judgments is partially determined by the overlap in social contexts between perceivers. We examined this aspect of interpersonal perception by using social network analysis to measure perceptions of personality across different social contexts. Participants (N=52) completed the IPIP-NEO Five Factor Model (FFM) inventory, describing their personalities as they generally see themselves. They then constructed an ego-centered social network, describing the context and quality of their own relationships with 30 acquaintances, and the relationship between each pair of acquaintances. Finally, participants completed a brief dyadic measure of the FFM, describing their own personalities when relating with each acquaintance. Up to six informants were selected from differing regions of each participant’s social network, and each informant completed the IPIP-NEO about the target participant (N=228). Linear mixed model analyses indicate that variance in informant judgments is best accounted for by a combination of the global perception of one’s personality, characteristics of the social network, and idiosyncratic aspects of the specific relationships. Results suggest that a social network approach to assessment may be a useful way to capture variability in interpersonal perception.

11. Denial and Undercontrol are Related to Externalizing Behavior Problems in Early Adolescence

Phebe Cramer, Williams College

Hierarchical Linear Modeling was used to trace the developmental trajectory of Undercontrol and Externalizing Behavior Problems from early childhood through early adolescence, using longitudinal data from the Institute of Human Development, University of California, Berkeley. At early childhood, late childhood, and early adolescence (a) Undercontrol was assessed using Q-sort data; (b) Externalizing Behavior Problems were assessed on the basis of mothers’ reports of behavior problems. At early adolescence, Denial was assessed from TAT stories. It was predicted that early adolescent use of the immature defense mechanism of Denial would be related to other signs of immaturity – namely, ego undercontrol and the presence of externalizing behavior problems. Results indicated both Undercontrol and Externalizing Behavior were significantly related to the use of Denial at early adolescence. However, Undercontrol was related to Externalizing Problems only for those adolescents who were strong users of Denial. Analysis of the developmental trajectories (linear slopes) of the two variables showed that there was variability in the direction of change in Undercontrol and Externalizing Problems; the majority of children decreased, but some increased. For the children who increased in Undercontrol, the use of Denial was related to magnitude of increase (linear slope); for the children who decreased, Denial was not related to magnitude of change. For children who increased in Externalizing Behavior, Externalizing scores at all three ages (early childhood, late childhood, early adolescence) were related to Denial. For children who decreased in Externalizing Behavior, this relation was not found. Thus, an increase in Undercontrol from early childhood to early adolescence predicted the use of Denial at early adolescence, whereas for children who increased in Behavior Problems there was a positive relation between Externalizing Behavior at each age and the use of Denial at early adolescence.

12. Psycho-lexical Openness to Experience

Boele De Raad and Dick Barelds, University of Groningen, The Netherlands

While McCrae (1990) argued that Openness to Experience is inadequately represented in natural language trait terms, the present study (in Dutch) shows that there is no problem whatsoever to find a sufficient number of descriptors in the natural language to reliably define both the domain of Openness to Experience and the corresponding six facets. From a non-restrictive list of 2,331 trait descriptors (Barelds & De Raad, 2008), 125 items were selected meeting the definitions of NEO-PI-R Openness to Experience and its facets. 87 (the sample is being enlarged) participants filled out the lexically derived Openness-and-facet-scales, the NEO-PI-R Openness-and-facet-scales, the Dutch Five Factor Personality Inventory, and eight scales measuring the recently published eight-factor model of personality. In terms of internal consistency, the lexical Openness-scales outperformed the NEO Openness-scales. Corresponding scales from the two Openness instruments generally correlated higher than non-corresponding scales. The study by De Raad and Barelds (2008) did however, not give rise to a separate Openness factor. Instead, its semantics were distributed in various niches of the eight-factor model. The present data generally support the latter view. Separate factor-analyses of the two sets of Openness items (NEO-PI-R & lexical), though supportive of the six facets, consistently confirmed a two-factor solution describing Openness to internal and to external experiences (cf. Griffin & Hesketh, 2004). Correlations with other trait scales are provided.

13. Partner Preferences Among The Gifted

Pieternel Dijkstra, Dick P.H. Barelds, Noks Nauta, and Sieuwke Ronner, University of Groningen, The Netherlands

To date hardly anything is known about the partner preferences of gifted people. The present study therefore examined the partner preferences of 129 single and gifted (IQ > 130) individuals (mean age = 37.43, SD = 12.14). Gifted people found it most (and about equally) important that a potential partner would be kind, understanding, intelligent and mentally healthy. Gifted men, more than gifted women, valued good genes in a potential partner whereas gifted women, more than gifted men, valued good mental health and solid earning capacities in a potential partner. In addition, following the work of Dijkstra and Barelds (2008), the present study examined the extent to which gifted people desired a complementary or a similar partner in terms of personality. When explicitly asked, 46% of the participants reported finding it (very) important that a mate possessed similar personality characteristics. When asked to rate their own and their ideal partner’s personality characteristics (in terms of the Five-Factor Model) findings showed a remarkably strong similarity between self-rated personality and the personality characteristics desired in a mate (r’s for the total sample ranging from .35 for Neuroticism to .70 for Conscientiousness, p’s < .001) This was not only true for those 46% of the participants who judged similarity in personality to be important, but also for those who reported finding personality similarity not important. A possible explanation is that, when looking for a mate, people often lack the introspective awareness of what they desire in a mate. To increase the comparability of the present study’s findings, the present study is currently being cross-validated in a community sample of non-gifted people.

14. Gender Differences in Claimed Self-Handicapping: The Value of Effort Scale

Josh Eblin and Robert Arkin, The Ohio State University

The aim of this research is to shed light on the underlying causes of the oft-observed gender differences in self-handicapping behavior. McCrea and colleagues in two sets of studies (2008) propose that women value effort more than men, and argue further that this difference in values explains the gender difference in self-handicapping. We agree, but add that this difference in values is likely due to a greater “moralizing” of the exertion of effort among women than men. A chief purpose of the studies reported here is to begin the process of validating the Value of Effort Scale, which was designed to shed light on this gender difference. Study 1 was designed to explore when individuals will claim a handicap, and whether or not gender moderates this behavior. Based on prior research, it was expected that women will not claim a self-handicap, while men will. Study 2 was designed to investigate why individuals will claim a handicap to their performance, and whether or not gender moderates this behavior. This hypothesis was tested by informing participants that they have self-handicapped on an experimental task, and then assessing their emotional life. It was expected that women’s affect (morally weighted and morally neutral) would become negative when told that they have self-handicapped, but that men’s affect would remain stable and unaffected. Implications of these findings are potentially broad. Self-handicapping has been found to be relevant to individual performance in academic and work settings as well as relevant to clinical conditions such as alcoholism. It has been fairly well established (Jones & Berglas, 1978; Berglas & Jones, 1978) that women are less likely to self-handicap behaviorally than are men, and this research points toward a fuller understanding of why this gender difference occurs.

15. From conscientiousness to life satisfaction: Decoding the mystery

Jennifer V. Fayard and Brent W. Roberts, University of Illinois at Urbana-Champaign

Richard W. Robins, University of California, Davis

Previous research has shown that the Big Five trait domain of conscientiousness is related to emotion, particularly to higher life satisfaction (DeNeve & Cooper, 1998). However, the mechanisms behind this connection have been largely unexplored. One possible explanation for this phenomenon is that conscientious individuals promote positive affect through accomplishments and having positive life experiences and avoid negative affect by avoiding negative experiences. In a four-year longitudinal study of undergraduates (N = 535), we examined this possibility using students’ cumulative GPA and evaluations of life events experienced while in school as potential mediators of the relation between conscientiousness and life satisfaction. In the first set of analyses, results confirmed that conscientiousness at year 1 predicted higher life satisfaction at year 4. Further, students’ cumulative GPA partially mediated the relationship between conscientiousness and life satisfaction. Conscientiousness also predicted experiencing more positive life events. While experiencing more positive life events predicted year-4 life satisfaction, positive event ratings did not mediate the relation between conscientiousness and life satisfaction. These results indicate that achieving a high GPA, but not experiencing more positive life events, could help explain, in part, the relation between conscientiousness and life satisfaction.

16. Precursors to Gender Attitudes in the Air Cadet Gliding Population

Emily-Ana Filardo, Angela Febbraro, Ritu Gill, Tara Holton, and Tonya Hendriks, Defence R & D Canada

Within the air cadet gliding population in Canada, females, who represent only 25% of the total population, are involved in 75% of the gliding accidents. While the reasons for this over-representation of females may be complex, what is clear is that females, in general, have faced negative attitudes and stereotypes regarding their place in the world of aviation. Stereotype threat, therefore, may play a role in the performance of female air cadets. This study investigated the precursors to negative gender-related attitudes in the air cadet gliding population. A structural equation model fitted to the data indicated that air cadets who had a rational thinking style and those who were likely to be involved in risky recreational activities were significantly more likely to have favorable attitudes towards female air cadets, compared to air cadets who performed risky health behaviors and who had an experiential thinking style. An experiential thinking style was also positively related to flight safety attitudes, which, in turn, were negatively related to attitudes towards female air cadets. Flight safety attitudes, a latent variable, was composed of five observable variables: flight emergency management attitudes, recognition of stress and fatigue effects (negatively related), recognition of accident susceptibility, perceptions of human factors relevance, and beliefs about pilot selection criteria. Positivity towards female air cadets, also a latent variable, was composed of four observable variables: beliefs about female pilots' flight proficiency and flight confidence (both positively related), and beliefs about females' safety orientation (i.e., over-cautiousness) and flight standards that unfairly favored females (both negatively related). It is suggested that performance pressure on females may result from negative attitudes towards female pilots as theorized in the stereotype threat literature, and could contribute to the disproportionate number of gliding accidents involving females.

17. Set-point change and adaptation after the birth of the first child

Judith Gere and Ulrich Schimmack, University of Toronto

Adaptation theory of well-being proposes that most events affect well-being only temporarily and happiness returns to a set-point level (Brickman & Campbell, 1971). However, there are individual differences in set points and adaptation and some events can lead to a change in set-points (Diener, Lucas, & Scollon, 2006). The goal of this study was to examine set-point and adaptation theory with birth of the first child as a life-changing event. The sample included 6455 participants (3137 male, 3317 female) from the German Socio-Economic Panel Study who reported a birth of a child during a yearly interview, and who had reported life satisfaction annually starting from four years prior to childbirth to four years after childbirth. A two-intercept, two slope latent growth curve model was used to test set-point changes and adaptation effects. Examination of the mean slopes indicated that life satisfaction gradually increases up to childbirth, reaching a peak in the year of the childbirth, and then decreases in the years following childbirth. Gender moderated the effects after birth, which indicated that women on average had a stronger response to the event. A stronger adaptation effect for women showed that this effect does not last. This finding is consistent with other findings that average levels of men and women’s life satisfaction remain fairly stable (Diener et al., 2006). The most important finding was a high correlation between the intercepts before and after childbirth (r = .73), indicating that set-point before and after the birth of the first child remained fairly stable. However, stability was far from perfect, which suggests that the birth of a child changes life satisfaction over the first year after childbirth. Future research will examine potential moderators of this effect.

18. Abstracting and Extracting: Causal Coherence and the Development of the Life Story

Azriel Grysman, Rutgers University

Maintaining a causal relationship between past events and current notions of the self is crucial in integrating memories in the adult life story. Consensus is that the life story develops in adolescence, though little direct evidence from life story memories of pre-adolescents has been analyzed. This study compared episodic memories of emerging adults (age 18-22) and early adolescents (age 13-15) for life story events and other memories, in an attempt to distinguish characteristics of the life story. Participants were also asked to describe the connection between the stories told. Stories were analyzed for three measures of causal coherence: 1) meaning making, 2) narrative complexity, and 3) the use of causal terms. Results show an impact of age in two measures (meaning making and narrative complexity) and of story type (life story vs. non-life story) in all three. Effects of age show that young adults' narratives showed more evidence of self-related abstract thinking and the ability to see multiple dimensions. Effects of story type indicated that turning point narratives and event connections narratives contained more self- related lessons and insights, displayed greater recognition of complexity, and employed more causal terms. Descriptions of peak experiences and low points did not differ significantly from other episodic memories on these measures of coherence. Findings show that two important narrative characteristics, narrative complexity and reference to self-related lessons and insights develop in adolescence. These may be considered to be critical building blocks in the construction of the life story. Results also suggest that turning point narratives play a pivotal role in combining episodic memories to create the life story, a role that ought to be explored further.

19. Assessing the Impact of Combat Experience on Personality Change and the Development of Post-Traumatic Stress

Peter Harms, University of Nebraska-Lincoln

Cpt. Paul Lester, United States Military Academy

While the general developmental tendencies of individual differences have been well-established in psychological research little or no research has investigated the development of psychological characteristics in response to extreme environmental conditions. The Neo-socioanalytic theory of personality development postulates that personality changes occur in response to life experiences and the social expectations that come with assuming new life roles. In the present study, we use a longitudinal design to track the psychological development of over 600 military personnel as they train for and engage in active combat deployment. Personality was assessed using measures of Risk-taking, Courage, Propensity to Trust, Leadership Efficacy, Values, and Psychological Capacities. The first two waves of data were collected during combat training and final deployment preparation in Germany. Wave 3 was collected in Salman Pak, Iraq during combat operations. Wave 4 will be collected when active deployment ends. Initial analyses indicates that individuals higher on risk-taking were more likely to have engaged in direct and indirect fire engagements, to have been attacked by improvised explosive devices, and to have developed post-traumatic stress. For officers, the frequency of being attacked was associated with lower levels of leadership efficacy. Contact and support from family members was associated with higher levels of courage and having a purpose in life, in addition to lower levels of post- traumatic stress. Positive ratings of combat experience were associated with higher levels of leadership efficacy, psychological capital, and having a purpose in life. Overall, the current results demonstrate not only the importance of individual differences in determining actual combat experience, but also the psychological experience of combat experience. Moreover, the present study offers insights into the both the negative effects of traumatic events and the positive role of social support networks on the development of values, attitudes, and psychological disorders.

20. Perfectionism and Suicidal Risk in a College Student Population: Does Loneliness Affect the Link?

Kathleen E. Hazlett, University of Michigan
Jameson K. Hirsch, Eastern Tennessee State University
Edward C. Chang, William Tsai, Kavita Srivastava and Jean M. Kin, University of Michigan
Elizabeth L. Jeglic, John Jay College of Criminal Justice
Ratika Singh and Melissa Ng, University of Michigan
Lawrence J. Sanna, University of North Carolina

Findings from research over the past three decades have shown that perfectionism is often associated with greater maladjustment. For example, studies using the Frost Multidimensional Perfectionism Scale have shown that greater perfectionism is associated with greater depressive symptoms and suicidal risk. The FMPS breaks perfectionism into six dimensions, namely, concern over mistake, personal standards, parental expectations, parental criticism, doubts about actions, and organization. Among these dimensions, concern over mistakes and doubts about actions have been found to be most frequently associated with maladaptive outcomes. With regards to maladjustment, although some studies have shown a link between perfectionism and suicidal risk much of the literature has focused on the prediction of psychological symptoms. Thus, numerous studies have now pointed to a reliable link between perfectionism and depressive symptoms. Although such findings indicating a direct link between perfectionism and maladjustment are interesting, a growing number of researchers have focused on identifying additional factors that may minimize or maximize this link. Indeed, in a recent study, Chang et al. found that the presence of loneliness may intensify the association between perfectionism and psychological symptoms in college students. That is, these researchers found evidence for loneliness as a moderator of the link between perfectionism and psychological symptoms. To expand on these findings, the present study sought to examine the impact of loneliness on the relationships between perfectionism and variables implicated in suicidal risk, namely hopelessness, past suicidal behaviors and self-harm, within a sample of 386 college students. Results based on correlational and regression analyses indicated that loneliness significantly impacts the relationships between perfectionism and these variables. These relationships are evident for some, but not all, dimensions of perfectionism. Implications of findings will be discussed.

21. The Dynamics of Social Roles, Goals, and Personality States: A Bottom-up Perspective

Daniel Heller, Tel Aviv University

Wei Qi Elaine Perunovic, University of New Brunswick

Daniel Reichman, Tel Aviv University

Recently, we theorized a bottom-up model of personality in which traits can develop and change from the accumulation of daily situations and behaviors over time (Heller, Komar, & Lee, 2007; Heller, Perunovic, & Reichman, in press). We posited that roles, which represent important classes of situations, can elicit different types of short-term goals. We then argued that these goals can serve as psychological components of situations, thus exerting an influence on personality states, which aggregated over the long-term can shape broad personality traits. In this paper, we test the first two predicted links via two diary studies (N=100 and N=74) in which Canadian undergraduates reported their short-term experiences (i.e., in the last two hours) repeatedly (i.e., three times a day) for 10 days, including their: a) personality states, b) short-term goals, and c) social roles they occupied. First, we observed important differences in goal pursuit as a function of role shifts. Specifically, participants reported pursuing more intrinsic and less extrinsic (thus more self-concordant) goals during the times they occupied a friend role, relative to a student one. Second, important within-individual links between goals and personality states were found. Consistent with predictions, participants reported higher levels of state neuroticism when pursuing extrinsic goals, and lower levels when pursuing intrinsic or self-concordant goals. In contrast, participants reported higher levels of state extraversion when pursuing intrinsic or self-concordant goals, and lower levels when pursuing extrinsic goals. This is the first paper to examine the within-individual processes that occur naturally in people’s daily lives pertaining to the covariation between roles, goals, and personality states. However, because the available empirical evidence is correlational at this stage, it remains for future experimental research--in which participants’ roles or short-term goals are manipulated repeatedly and subsequent goals or personality states are observed--to demonstrate the precise causal sequence.

22. Motives, Abilities, and Perceptions Underlying the Dimensions of Extraversion

Molly Hensler and Dustin Wood, Wake Forest University

Although considerable attention has been given to how personality traits relate to variation in behavior, it is not well understood what causes variation in personality traits. In two studies, we thus attempt to identify the motives, abilities, and perceptual tendencies that underlie variation in extraversion. First, students completed a measure of their Big Five personality traits, and participants who scored at the extremes of extraversion then completed a follow-up interview where we attempted to elicit statements about the motives, abilities, and perceptions that caused them to regularly act introverted or extraverted. Second, another sample of students was asked to recall people they knew who were highly introverted or extraverted, and describe the motives, abilities, or perceptual tendencies that might lead those people to behave this way. Both sources of data were coded to identify themes that were consistently provided to explain why people engaged in introverted or extraverted behavior. Results showed that extraverts typically explained their extraverted behavior as emanating from a desire to meet other people. Extraverts tended to report behaving in extraverted ways because they were comfortable around others, liked to meet people, enjoyed learning things about those people, and enjoyed connecting with others. In contrast, introverts explained their behavior as emanating from heightened concern about how others would perceive them. Many admitted they were untalkative because they felt they were inferior to others, did not think others were trustworthy, thought they might be bothering others, or feared that others would perceive them as awkward or think they sounded stupid if they spoke. The results provide clues to the motives and cognitions underlying extraverted and introverted behavior, and to points of intervention for efforts to change maladaptive extraversion levels.

23. Unraveling the three faces of self-esteem: A new information-processing sociometer perspective

Sarah Hirschmüller,Mitja D. Back, University of Leipzig, Germany

Sascha Krause, Westfälische Wilhelms-University Münster, Germany

Boris Egloff, University of Leipzig, Germany

Stefan C. Schmukle, Westfälische Wilhelms-University Münster, Germany

Based on an integration of sociometer theory and information-processing models, the present study investigated the predictive validity of three self-esteem measures: self-report, an implicit association test, and an affective priming task. In a first session, self-esteem measures were obtained from 93 participants. After an interval of four weeks, interpersonal perception ratings were collected in small round-robin groups. Participants were requested to briefly introduce themselves to the group before evaluating one another and indicating how they expected to be evaluated by the others (meta-perceptions). As hypothesized, all three self-esteem measures independently predicted the perception of being valued (PBV) in a real-life situation. In sum, the present study shows that three independent faces of self-esteem can fruitfully be distinguished, a finding that has important implications for the measurement and understanding of self-esteem.

24. Exploring the Links between Trait Structure and Social-Cognitive Processes: The Case of Personality Vulnerabilities to Psychopathology

Ryan Y. Hong, National University of Singapore

Sampo V. Paunonen, University of Western Ontario

The present research argues that integrating two parallel approaches to personality-psychopathology relations (i.e., dispositional trait and social-cognitive) is crucial for advancing current understanding on the etiology of a wide range of psychopathology, including depression, anxiety, and substance use. Study 1 explored the relations among various social-cognitive vulnerability variables (i.e., depressogenic inferential style, dysfunctional attitudes, rumination, anxiety sensitivity, intolerance of uncertainty, social-phobic inferential style, poor self- control/regulation) and found that they could be organized into a two-factor model representing general vulnerability factors to internalizing and externalizing psychopathology, respectively. Furthermore, these two vulnerability factors showed differential linkages with the Big Five traits of Neuroticism, Extraversion, and Conscientiousness. processes. Notable findings include: (a) daily social-cognitive processes covaried within individuals; (b) these Study 2 employed a longitudinal experience-sampling design to examine participants’ day-to-day social-cognitive processes were related to Neuroticism; and (c) these processes mediated the relations between Neuroticism and daily anxiety/depression symptoms. Overall, these data suggest the need to explore further the possible links between the structural and processing aspects of personality vulnerabilities to psychopathology.

25. The Impact of Anime/Manga on Personality Development of Youth

Akihiko Ieshima, Kyoto University

The concern with Japanese comics (so-called “manga”) in academic fields has been growing and some psychological studies on manga and manga fans have been made over the past few years in Japan. As early as 1944, Bender asserted the important role of comics on development of children. After pointing out that the art form and language of the comics represent experimentation for the child, Bender (1944) goes on to say: “The positive qualities of the comics are their adaptability and fluidity in dealing with social and cultural problems, continuity and repetition, and an experimental attitude and technique.” My research interest has been in the development of narrative identity throughout the life-span. In the past few years, I have been examining the impact of manga on the process of identity formation by means of questionnaires and in-depth interviews. In my study, Over 300 university students completed the questionnaire form and over 30 young people participated in the interview. In short, the results of these studies suggest that some Japanese people, especially those who are exposed to manga since youth, learn more about their ideal-self, moral sense, knowledge and behavior from fictitious characters (e.g., manga heroes and heroines) than from people in their immediate environment (e.g., parents, teachers, friends, etc.). It was found from the results when and how Japanese young people came to read manga. Basically most of them read manga only to get relaxation but they learn a lot from manga after all. It was also found from online survey that reading manga gives people motivation, relaxation, and self-reflection.

26. Regulation and Personality mechanisms of decision making in emergency situations.

Tatiana Indina and V. Morosanova, Psychological Institute of Russian Academy of Education, Russia Moscow

Subject of the study: Self-regulation and personality factors of decision making in lifesavers professional activity. Sample: 100 lifesavers of Moscow emergency situations department. Methods: NEO PI-R (V.Oryol, I.Senin adaptation), Self-regulation profile questionnaire (V.I Morosanova), Personality factors of decision making questionnaire (T.V. Kornilova), Decision making model (T.Indina). To study the effectiveness of decision making in emergency situations specific experimental model was elaborated. A number of professional tasks were worked out to diagnose main decision making parameters: search for information, situation assessment, subjective task separation, alternatives construction, choice of the alternative, decision implementation.

Results and conclusions: NEO PI-R Personality Scales (Openness to Experience, Agreeableness, Conscientiousness) as well as Individual self regulation basic components (goal planning, programming of actions, modeling of significant conditions, result estimation) have shown significant positive correlations with decision making effectiveness. It was proved that individual self regulation development and certain personality domains improve decision making effectiveness in emergency situations.

27. Individuals with the Temperamental Trait of Sensory Processing Sensitivity Notice Subtleties: Neural Response to Changes in Visual Scenes

Jadzia Jagiellowicz, Xiaomeng Xu, Arthur Aron, Elaine Aron, Stony Brook, University New York

Guikang Cao, Tingyong Feng, Southwest University, China

Xuchu Weng, Chinese Academy of Sciences, China

This study examines the extent to which individual differences in the adult temperamental trait of sensory processing sensitivity (SPS) correlate with neural activation in response to subtle changes in visual stimuli. SPS is an adult temperamental trait, moderately correlated with introversion and neuroticism (Aron & Aron, 1997), and characterized by sensitivity to both internal and external stimuli. Sixteen Chinese students aged 19 to 25 years (M = 21.6, SD= 1.63) completed the Highly Sensitive Person (HSP) Questionnaire. Subsequently, participants were asked to detect minor and major changes in both slowly and quickly presented scenes while undergoing functional magnetic resonance imaging. Individuals high in SPS evidenced greater intensity of brain activation when detecting minor changes in stimuli than did individuals low in the trait. Activation was found in brain areas involved in high order visual processing and in attention, as well as in oculomotor control. Strong and significant correlations were found in the claustrum (r =.89), in two areas in the middle temporal gyrus, (r =.84 and r =.83), in the sub-gyral temporal lobe (r= .85), and in the cerebellum (r= .81). Correlations in these regions remained strong and significant after partialling out neuroticism and introversion. After controlling for neuroticism and introversion, region of interest analyses yielded correlations (ranging from .62 to .84) in the right hemisphere as follows: the temporoparietal junction, the intraparietal sulcus, and the middle frontal gyrus. These findings provide the first evidence of neural differences in SPS and provide preliminary evidence for heightened sensory processing in individuals high in SPS.

28. Calibrating Personality Self-Report Scores to Acquaintance Ratings

John A. Johnson, Pennsylvania State University, DuBois

By convention in individual personality assessment, scores on self-report questionnaires within ±.5 or ±1 standard deviation of the mean score for that trait are considered "average," whereas scores outside that range are reported as "high" or "low" levels of the trait. To date, no one has examined how well this convention corresponds to perceptions of trait levels by acquaintances. The present research demonstrates exactly what range of scores on a self-report personality inventory correspond to acquaintances' perceptions of low, average, and high trait levels. 151 research participants completed Goldberg’s (1999) 300-item International Personality Item Pool (IPIP) representation of Costa and McCrae’s (1992) NEO PI-R (hereafter, IPIP-NEO). Three knowledgeable acquaintances of each participant were sent to a Web site containing descriptions of the five domains and 30 facets measured by the IPIP-NEO. Acquaintances were asked to consider the participant’s trait level compared to other persons in the population of the same sex and roughly the same age. They rated participants on 35 scales with the following percentile anchor points: 1, 10, 20, 30, 40, 50, 60, 70, 80, 90, 99. The anchors were grouped into sections labeled as follows: 1, 10: very low (lowest 15% of the population), 20: low-average (16-29%), 30 through 70: middle-average (30-70%), 80: high-average, (71-84%) and 90, 99 very high (85-99%). The middle percentile category corresponds to ±.5 SD from the mean under normality, the low-average and high-average to -1 and +1 SD from the mean, respectively, and the remaining categories, more than 1 SD from the mean (Anastasi, 1976). Plots of standardized IPIP-NEO domain and facet self-report scores against the acquaintance percentile ratings show how often traditional “low,” “average,” and “high” self-report scores actually correspond to acquaintances’ perceptions of those levels.

29. Belief in a just world and blaming the victim

Kaoru Kurosawa, Nozomi Doi, and Miho Shirai, Toyo University

According to Just World Hypothesis, people more or less believe that the world is essentially fair, and that the good people are rewarded and the bad punished. It implies that a victim of random street assaults, for example, might be blamed for his/her part in the incident. However, predictions by the hypothesis are by no means clear, and presence of offender, for instance, might possibly complicate the picture. If the high believers know the criminal party, do they still blame the victim, not the offender? We investigated the phenomenon with the salience of victim and offender manipulated. First, we developed a new Japanese version of JW Scale, more adequate for empirical research than before. We constructed a 10-item scale, with an alpha of .74, which we judged to have one factor. Participants in the main study read a newspaper article about a street crime, and responded to a series of rating scales: seriousness of the crime, haphazardness, the degree of victim's fault, and so on. In addition, in offender salience condition (N=48), a short paragraph was added at the end, citing a comment by someone who knew the offender. In the victim salience condition (N=45), it was a comment about the victim. In control condition (N=45), there was no comment. Overall results indicated that JW score had a positive correlation with the degree of carelessness on the victim's part (r=.271, d.f.=136); the stronger the JW belief, the more careless the victim was judged. It also had a negative correlation with the role of luck in the incident (r=-.229); the stronger the belief, the less important the victim's luck was judged. However, the correlation between the JW score and the degree of victim's fault was not statistically significant. More complete results will be discussed, with structural equation analysis of the data.

30. Normative assumptions underlying the DSM-IV personality disorder criteria

Daniel Leising, Martin Luther University, Halle-Wittenberg, Germany
Julia Ostner, Georg August University, Göttingen, Germany

Kate Rogers, Wake Forest University, NC, USA

Clinical diagnoses are impossible without referring to implicit or explicit assumptions about desirable functioning. The normative assumptions underlying the DSM-IV personality disorder (PD) criteria are largely unclear. We conducted a small scale empirical study in which we first explicated these assumptions by logically "inverting" each of the PD criteria, and then cluster-analyzing them based on similarity ratings. Thereby we obtained a hierarchical structure of 10 higher-order (e.g. "Be ready to take risks", "Trust other people", "Control yourself") and 27 lower-order imperatives regarding desirable behavior. Such imperatives may be justified by referring to different frameworks: (1) a given psychotherapist's personal value system (2) the expectations of the culture in which a person currently lives (3) the expectations of the culture in which a person was raised (4) generalized assumptions about "natural" personality functioning that are rooted in evolution theory (5) the presence of distress or impairment. Of these, we argue that the evolutionary framework is the most convincing. Only within this framework it would be reasonable to investigate the biological underpinnings of personality disorders. The distress / impairment framework deserves further clarification as to (a) whose suffering is relevant (only the patient's or others' as well?) (b) what distinguishes everyday suffering from pathological suffering? (c) should diagnoses require that suffering /impairment is currently present, or would it be sufficient if suffering /impairment is highly likely in the future? (d) who is authorized to judge the presence of suffering / impairment (e.g. what if a patient denies being impaired?). Future editions of DSM should aim to define personality pathology in less culture-relative terms.

31. Mechanisms of Personality Trait Change in Older Adulthood

Jennifer Lodi-Smith, Center for BrainHealth, University of Texas at Dallas

Brent. W. Roberts, University of Illinois, Urbana-Champaign

Jacqui Smith, University of Michigan

Recent findings suggest that changes in personality traits during older adulthood are an important public health issue (i.e. Mroczek & Spiro, 2007). Because changes in personality traits have a considerable impact on the health of older adult populations, understanding the mechanisms underlying personality trait change during older adulthood is critical to developing means of preventing harmful declines in personality traits. Using data from the Health and Aging Study of Central Illinois (HASCI) and the Berlin Aging Study (BASE), the present research examines how two significant shifts associated with aging – (1) changing social roles and (2) changing cognitive functioning – impact personality trait change after age 60. For the 100 older adults assessed in the HASCI data, being involved in romantic relationships and being committed to community activities such as church or volunteerism in 2002 prevented declines in agreeableness, conscientiousness, and emotional stability between 2002 and 2007. In addition, for the 206 seniors who participated in BASE between 1990 and 1996, declines in speed of processing corresponded to declines in extraversion and emotional stability during that same period. The evidence presented in the current research suggests that the shifting worlds of seniors trigger personality trait change. The present findings provide some initial road marks for how we may be able to foster a healthier senior population by promoting the maintenance of a healthy personality trait profile through social and cognitive stimulation.

32. Revolutions, coups, and clashes: Predicting civil unrest through analyses of implicit motives in political speeches

Alanna Maguire, B.A., Sara Konrath, PhD., University of Michigan

Political psychologists hypothesize that war is more likely to break out in times when power motives in leaders are high and affiliation levels are low (Winter, 1993), however these analyses have been limited to conflicts between individualist countries. We examine four revolutionary movements in a collectivist country (the Philippines) to examine whether this pattern applies to political violence in general, including in intra-national conflicts. Speeches made by government officials and opposition leaders were gathered and scored for power, affiliation, and achievement images. We find the highest power, lowest affiliation, and lowest achievement in the most violent conflict.

33. Lifetime exposure to narrative fiction predicts recognition of facial emotion

Raymond A. Mar, Ph.D. and Taras Babyuk, York University

Reading narrative fiction appears to entail a cognitive and emotional simulation of social experience (Mar & Oatley, 2008), and individuals who frequently engage in this type of reading may thus bolster or maintain their social abilities. Consistent with this proposal, previous research (Mar et al., 2006) has demonstrated that lifetime exposure to narrative fiction is positively correlated with the ability to infer mental states based on pictures of a person’s eye region (i.e., the Reading-the-Mind-in-the-Eyes Task; Baron-Cohen et al., 2001) and to decode social information from nonverbal cues (i.e., the Interpersonal Perception Task-15; Costanzo & Archer, 1993). The current study examines whether exposure to narrative fiction relates to one’s ability to recognize brief displays of facial emotion. Participants (N = 260; 182 female) completed a measure of lifetime exposure to print (Author Recognition Test-Revised; Mar et al., 2006) and viewed pictures of faces displaying emotions. Happy, sad, angry and surprised faces were shown for 1200 msec, after which participants were asked to label the face as belonging to one of those four categories. Six faces were shown for each emotion, and the order of presentation was randomly determined. Controlling for age, gender, and years of English fluency, exposure to narrative fiction predicted better recognition of facial emotion (Standarized Beta = .11, P < .05 [one-tailed]). The same, however, could not be said for exposure to expository nonfiction (Standarized Beta = .02, P > .05 [one-tailed]). Although this study is correlational and direction of causality can thus not be inferred, it contributes to a growing body of evidence that many narrative media have social components, such as recent work on anthropomorphization (e.g., Gardner & Knowles, 2008).

34. Inspiration and the creativity of writing: Person, process, and product

Laura A. Maruskin, Scott E. Cassidy, Todd M. Thrash, College of William and Mary

Within the creativity literature, inspiration has been the subject of much discussion but little research. We examined the role of inspiration in the writing process. Our goals were to (a) establish a nomological network of the inspiration construct, (b) document the utility of inspiration in predicting judges’ ratings of creativity, and (c) test a theoretical model of traits as antecedents and moderators of the relation between inspiration and creativity. One-hundred sixty-two undergraduates completed trait questionnaires and were given the opening paragraph of a mystery story. After generating an idea for how to complete the story, participants appraised the creativity of their idea and completed measures of inspiration, effort, and positive affect. Participants then completed the story on a computer, and screen snapshots were recorded throughout the writing process. Snapshot data were used to compute indexes of output (number of retained words), efficiency (proportion of typed words that were retained), and productivity (number of retained words per minute of writing). Ten judges coded participants’ stories for creativity and mechanics following Amabile’s guidelines. Regarding our first goal, inspiration was related to greater output, efficiency, and productivity, thus establishing a nomological network of objective variables. Regarding our second goal, inspiration predicted judges’ ratings of creativity (but not mechanics), even when alternative predictors (effort, positive affect, verbal SAT scores, openness, and approach temperament) were controlled. Regarding our third goal, inspiration (but not effort or positive affect) mediated between creativity of the idea and creativity of the product. Openness functioned as an antecedent of creativity of the idea, whereas approach temperament moderated (amplified) the relation between creativity of the idea and inspiration. These findings integrate the inspiration, creativity, and personality literatures.

35. Is Isomorphic Scaling of Personality Constructs Possible?

Robert E. McGrath, Fairleigh Dickinson University

Psychologists evaluate the quality of their measures through tests of construct validity. These tests usually focus on some aspect of item or scale behavior. Examples include using factor analysis to evaluate structural relationships between items, or tests of convergent and discriminant validity for the scale as a whole. Measurement theory in mathematics focuses instead on maximizing the function that maps locations on an attribute to the numeric scale representing that attribute. Ideally, the relationship between the scale and locations on the attribute is isomorphic: each location on the attribute is represented by one value on the scale and each value on the scale represents one location. In practice this ideal is often impossible to achieve, but improvements in measurement methods reflect incremental progress towards the ideal. Mathematical measurement theory dominates in certain scientific disciplines, most notably physics and economics, but has little impact on measurement in the social sciences. From the perspective of mathematical measurement theory, psychological measurement practice focuses largely on unreliability as an impediment to achieving isomorphism, but this is not the only impediment possible. It is hypothesized that traditional measurement methods in psychology, which were developed for use with performance-based measures, actually compromise the potential for approximating isomorphism when applied to the measurement of self-perceptual variables. The proposed poster will summarize some of the consequences of measurement models that do not emphasize accurate mapping between attributes and scales. It will also describe an alternative model that potentially offers a more construct-accurate approach to the measurement of personality, and describes research currently underway to evaluate that model.

36. Conversational processes and life storytelling in dating couples

Kate C. McLean, Western Washington University

Monisha Pasupathi, University of Utah

Autobiographical storytelling has been theorized to be a critical process in the development of a meaning-filled life story (e.g., McLean et al., 2007), but empirical evidence for this is scant, particularly for meanings that are actually retained over time. In two studies we examined the way that autobiographical storytelling was related to the characteristics and retention of meanings about the shared personal memories. Our design centered on newly dating romantic couples in which one individual told the other an important personal memory, previously unshared with this partner. We focused on whether meanings about the memory initiated with the teller or the listener, concerned change or stability in the teller’s self, the extent to which meanings were positive, the extent to which meanings became shared by both conversational partners, and whether meanings were retained by the teller at a 1-month follow-up assessment. Across two studies (n = 62 pairs in study 1; n = 68 pairs in study 2) we found that meanings generated in conversations tend to be positive, stability focused, and temporary both in terms of their failure to become shared (across individuals) and their failure to be retained (within one person). When meanings initiated with the listener, they were particularly less likely to be retained, and they were particularly less likely to be about change in the teller. However, listeners have an important role in the individual’s retention of meaning as, across both studies, meaning that was shared by both listener and speaker after their conversation was more likely to be retained. Discussion focuses on the role of important others and conversational processes in shaping our life stories.

37. Behavioral genetic models of temperament: Heritability, rating bias, and sibling contrasts

Paula Y. Mullineaux, Kirby Deater-Deckard, Virginia Polytechnic and State University

Stephen A. Petrill, Ohio State University

Lee A. Thompson, Case Western Reserve University

Laura S. DeThorne, University of Illinois

Individual differences in child temperament arise from complex genetic and environmental processes. Few studies have examined whether parents’ ratings of child temperament are best explained by common genetic and environmental influences or by specific parent rating biases during middle childhood. Data on the Child Behavior Questionnaire-Short Form (Putnam & Rothbart, 2006) were available for 88 MZ and 109 same-sex DZ twin pairs. Univariate analyses of Effortful Control indicated significant genetic and environmental effects across parents’ ratings. For Negative Affectivity, mothers’ ratings were consistent across the scales indicating significant genetic and environmental effects, whereas fathers’ ratings indicated negligible genetic effects for the Negative Affectivity factor, Discomfort scale, and Sadness scale. Sibling contrast effects were indicated for the Discomfort, Impulsivity, and Shyness scale for mothers’ ratings. Evidence of sibling contrast and dominance effects were suggested for the High-Intensity Pleasure and Impulsivity scales for fathers’ ratings. The psychometric and rater bias model for parental ratings of the highest-order factors also were examined. Although both models fit the data well, the rater bias model resulted in the best overall fit for Effortful Control (X2 (13) = 9.69, p = .72, AIC = -16.31, RMSEA = .00) and Negative Affectivity (X2 (13) = 5.20, p = .97, AIC = -20.80, RMSEA = .00). These results suggest that although for most aspects of temperament parents are similarly influenced by genetic and environmental effects, each parental rating also is influenced by their own ratings biases and differences in ratings may not simply reflect a different view or experience with the child.

38. Teachers’ assessents of children’s personality traits predict directly observed behaviors forty years later

Christopher S. Nave, Ryne A. Sherman, David C. Funder, University of California, Riverside

Sarah E. Hampson, Oregon Research Institute and University of Surrey, Guildford, UK

Lewis R. Goldberg, Oregon Research Institute

To the degree that psychology is the study of behavior, the ultimate outcome of interest is what people do. The current study analyzes data from the Hawaii Personality and Health cohort that includes teacher assessments of children in Hawaii made between 1959 and 1967 when the children were in Grades 1,2,5, or 6. Forty plus years later, follow-up health assessments and videotaped personality interviews were obtained from approximately 300 participants. The videotapes were coded for directly observed behaviors by four research assistants using the Riverside Behavioral Q-Sort (Furr, Funder, & Colvin, 2000). Analyses based on the current sample (N = 48) of the ongoing project indicates that teacher’s assessments of various children’s personality traits predict a number of directly observed behaviors years later, particularly with respect to the Big Five component of Openness/Intellect and several individual trait items (e.g., assertiveness, eccentricity, spitefulness, submissiveness).

For example, children rated high in assertiveness by their teachers were seen, years later, to have others seek advice from them (r = .44), to express sexual interest (r = .39), and to dominate the situation(r = .37), compared to children rated low on assertiveness. Children rated high in submissiveness were seen, years later, to express self-pity or feelings of victimization (r = .49), to express sympathy (r = .41), and not to regard themselves as physically attractive (r = -.48) or speak in a loud voice (r = -.41). Children rated high in eccentricity were seen, years later, to initiate humor (r = .53) and to show interest in intellectual or cognitive matters (r = .44), but not to seem to enjoy the situation (r = -.47) or appear relaxed and comfortable (r = - .47). This study may be the first to show the predictability of directly observed behavior from personality traits assessed decades earlier.

39. High Stability and High Variability in Personality Validated in Observer Reports

Erik Noftle and William Fleeson, Wake Forest University

The density distributions approach revealed that individuals’ Big Five personality states were both highly stable and highly variable in everyday life (Fleeson, 2001). Specifically, participants’ behavior averages were remarkably stable from one week to another, with correlations around .80-.90. At the same time, during the period of a week, most participants reported quite variable behavior, ranging from the lowest extreme of each trait to the highest. However, these findings were limited by their exclusive useage of self-reports of behavior, and behavior assessed in unmonitored everyday life contexts. The present studies extended previous research in at least two ways. First, Big Five state ratings were collected within a set of standardized laboratory situations. Thus, high stability in behavior averages, if found, could not result from individuals being in similar situations within each period, which might have been true of everyday life. Second, previous findings of stability and variability were tested using observer ratings of behavior, bypassing most self-report biases. Two laboratory-based studies employed experience-sampling, in which targets and observers rated behavior using Big Five adjectives. In Study 1, 44 targets (and one observer per target) rated the target’s behavior twice an activity during ten 60-minute activities (held across two months). In Study 2, 97 targets (and two observers per target) rated behavior during twenty 40-minute activities (held across a month). Means and standard deviations were calculated to assess individuals’ behavior averages and variabilities. Results for observer reports were quite similar to those for self-reports: substantial intraindividual variability was revealed in behavior, more than the variability between people. Behavior average stabilities across each half of the reports were high, suggesting stability is at least partly due to personality, and not just situation similarity.

40. Acquaintance Ratings for the Impostor Phenomenon

Julie K. Norem and Jonathan M. Cheek, Wellesley College

People who are usually successful yet doubt that they deserve success and worry that others will eventually perceive that they are “fakes” are said to be experiencing the “impostor phenomenon” (Clance, 1985). This psychological construct has received considerable attention in the popular press and some notice in the psychotherapy literature, but researchers have been skeptical about the validity of Clance’s Impostor Phenomenon Scale (IPS; e.g., Leary, Patton, Orlando, & Wagoner Funk, 2000). Particularly problematic is the finding that when asked to rate themselves and to indicate how they think other people regard them, high scorers on Clance’s IPS do not appear to believe that others view them more positively than they view themselves (replicated by McElwee & Yurak, 2007). We believe that availability of an acquaintance rating scale for the construct assessed by Clance’s self-report scale would facilitate progress in the debate about whether or not her scale actually measures genuine feelings of impostorism rather than a self-presentation strategy. Therefore we developed a set of items for acquaintances to use to rate college women who had completed Clance’s IPS (e.g., “She is more successful than she thinks she deserves to be” and “She is not as confident about her abilities as she really should be”). A seven-item version of this new acquaintance rating scale for the impostor phenomenon had an alpha coefficient of internal consistency reliability with two raters of .73 and correlated .43 (n = 110) with self-reports on Clance’s IPS; a nine-item version of the acquaintance rating scale had an alpha of .86 and correlated .40 (n = 47) with the IPS. These results suggest that the new rating scale wil be useful in future research on the validity of the imposter phenomenon.

41. Trait emotional intelligence: A scientific model of EI

K. V. Petrides, University College London (UCL)

The increasing number of faux intelligences (personal, social, emotional, practical, spiritual, creative, etc.) are characterized by conceptual confusion and, occasionally, shotgun empiricism intended to deflect attention from the fundamental theoretical problems of the various (often commercial) models. In the few cases where these models describe meaningful individual differences, these largely concern well-established personality traits. Trait emotional intelligence (trait EI or trait emotional self-efficacy) theory demonstrates how a realignment and enhancement of the higher-order personality dimensions (Giant Three or Big Five) can exhaustively accommodate all meaningful aspects of all EI-related models. The same principles (see, e.g., Petrides, in press) can be generalized to encompass other faux intelligences in order to incorporate them into mainstream differential psychology. In addition to the theoretical component of the paper, we present data from dozens of studies relating trait EI to criteria from a wide range of contexts, including organizational, educational, clinical, behavioral genetic, and experimental.

Petrides, K. V. (in press). Psychometric properties of the Trait Emotional Intelligence Questionnaire (TEIQue). In C. Stough, D.H. Saklofske, and J.D.A. Parker (eds.), Advances in the measurement of emotional inteligence. New York : Springer.

42. Development and Preliminary Validation of the Types of Intuition Scale (TIntS)

Jean E. Pretz, Illinois Wesleyan University

Jeffrey B. Brookings, Wittenberg University

Previous research has revealed that the construct of intuition is poorly understood and that the literature lacks a comprehensive measure of the construct. Factor analyses of current measures of intuition such as the experiential subscale of Epstein's Rational-Experiential Inventory and the intuitive-sensate subscale of the Myers-Briggs Type Indicator revealed that the two measures tap different aspects of intuition (Pretz & Totz, 2007). The experiential subscale measured preference for instincts, snap judgments, and gut feelings, but the intuitive-sensate subscale uniquely measured preference for abstract, holistic thought. The Types of Intuition Scale was developed as a more comprehensive measure of intuition in three aspects: holistic, inferential, and affective. Holistic intuitions are based on an integration of diverse and complex sources of information in a Gestalt-like and non-analytical manner. Inferential intuitions are the result of mental shortcuts and are based on previously-analytical processes which have become automatic. Affective intuitions are based on feelings. The Types of Intuition Scale is a 47-item questionnaire with 5-point Likert-scale responses that assesses preference for all three types of intuition in one instrument. The measure was administered to over 300 students and adults. Based on item and factor analyses, the subscales were condensed to a total of 36 items, and all three were found to have acceptable levels of reliability. Factor analyses supported the distinction among the three predicted types of intuition. The scales were validated by examining correlations with existing measures of intuition (Epstein's Rational-Experiential Inventory and the Myers-Briggs Type Indicator) and personality (Big Five and tolerance of ambiguity).

43. Low Self-Esteem Prospectively Predicts Depression

Richard W. Robins, University of California, Davis

Ulrich Orth, University of Bern

Background: Many theories of depression postulate that low self-esteem is a defining feature of depression. The two constructs are strongly correlated but the nature of their relation—specifically the causal direction—remains unclear. The present research uses longitudinal data to test three models. The vulnerability model hypothesizes that low self-esteem serves as a risk factor for depression; that is, negative beliefs about the self are not just symptomatic of depression but play a critical causal role in its etiology. The self-esteem buffering model hypothesizes an interaction between low self-esteem and stressful events; in the face of challenging life circumstances, individuals with low self-esteem are particularly prone to depression because they lack sufficient coping resources. The scar model hypothesizes that low self-esteem is an outcome rather than a cause of depression; episodes of depression leave “scars” in the individual’s self-concept that progressively chip away at self-esteem.

Method: The models were tested using data from several large longitudinal studies. Participants ranged in age from 15 to 96,

allowing us to examine whether the effects vary across the lifespan. All studies used the Rosenberg Self-Esteem Scale and either the Center for Epidemiological Studies-Depression Scale or the Beck Depression Inventory.

Results/Discussion: Cross-lagged regression analyses showed that low self-esteem predicted subsequent depressive symptoms (supporting the vulnerability model) but depressive symptoms did not predict subsequent levels of self-esteem (contrary to the scar model). This pattern of results replicated across studies, age groups, genders, affective/cognitive and somatic symptoms of depression, and after controlling for content overlap between the self-esteem and depression scales. Low self-esteem and stressful events independently predicted subsequent depression, but, contrary to the self-esteem buffering model, did not show an interactive effect. Future research should examine the processes through which low self-esteem contributes to depression.

44. Personality correlates of Attention-Deficit/Hyperactivity Disorder (ADHD) in Middle Childhood

Rasha A. Salib, University of Toronto

The present study extended the findings from Parker, Majeski, and Collin's (2004) research on adults by exploring the associations between the Five Factor Model of personality and ADHD in middle childhood. A large, moderately ethnically heterogeneous community sample of 250 children (age 7-12, 121 boys) was examined. Participating caregivers completed questionnaires gauging their children's personality and problem behaviour. Additionally, ADHD symptoms were assessed with a structured diagnostic interview completed by the caregiver. Data were analyzed using a series of hierarchical multiple regression analyses. Results were interpreted according to a dual-pathway framework, and they showed that higher- and lower-order personality traits were differentially correlated with symptoms. Specifically, total ADHD symptoms were negatively predicted by Conscientiousness and (un)Distractibility, and positively predicted by Extraversion, Activity level, and Openness. Further, inattentive symptoms after controlling for hyperactive/impulsive symptoms were negatively predicted by Conscientiousness, Intellect, (un)Distractibility, and Achievement Orientation, and positively predicted by Extraversion. Finally, hyperactive/impulsive symptoms after controlling for inattentive symptoms were negatively predicted by Compliance, Shyness, and Sociability, but they were positively predicted by Disagreeableness, Intellect, and Openness. Consistent with previous research using adult samples (Nigg et al., 2002) ADHD and inattentive symptoms were predominantly and strongly predicted by low Conscientiousness, whereas hyperactive/impulsive symptoms were primarily and strongly predicted by low Agreeableness. Findings support the theoretical connections between childhood personality traits and ADHD symptoms, and provide a novel insight into the associations between ADHD and lower-order childhood personality traits that were not previously examined. Overall, the differential pattern of results highlights the importance of identifying points of commonality and distinction between the ADHD subtypes.

45. Agency and Communion as indicators of personality in middle childhood.

Gregory C. Schell and Jennifer L. Tackett, University of Toronto

Agency and Communion have long been regarded as conceptual coordinates that organize and provide an underlying structure to personality and social interaction (Wiggins, 1991). In the narrative analysis literature, these concepts have been expressed as broad social motives served by a need for power and a need for intimacy (McAdams, 1980). This work has primarily focused on adults with little work done to examine how agency and communion influence the stories of children. The current investigation seeks to illustrate how the agentic and communal elements in children's stories are related to childhood personality. Children (114 boys and 116 girls) aged 9-10 (M=10.03, SD= .75) provided three imagined stories to stimulus cards from the Children's Apperception Test (CAT, Bellak, 1949). Each story was coded for descriptors of relevant agentic or communal behavior and were further categorized as positive/negative and attributed to the protagonist or other person. Children's personality was rated by mothers using the Inventory of Children's Individual Differences (ICID, Halverson, et al., 2003). The ICID is composed of five broad factors corresponding to the traditional Big Five factors and is further organized into 15 lower order facets. Correlations between CAT and ICID scores revealed that positive communal elements were associated with considerate, compliant, prosocial, positive affect and achievement orientation facets. Positive communal elements also predicted the higher order traits of extraversion, agreeableness and emotional stability. Additional results, including evidence for gender differences, are discussed in terms of the broader agency/communion theoretical framework. These findings suggest that themes of agency and communion are present in middle childhood and show theoretically-supported connections to the Big Five.

46. The Riverside Situational Q-Sort

Ryne A. Sherman, Christopher S. Nave, and David. C. Funder, University of California, Riverside

While a large number of psychological instruments measure global characteristics of persons (i.e. personality inventories), and at least some progress has been made in the development of instruments to measure global behaviors (e.g. Funder, Furr, & Colvin, 2000; Furr, in press), few instruments have been developed and validated to measure the psychologically important features of situations. This lack is surprising given years of social psychological research touting the power of situations (Haney, Banks, & Zimbardo, 1973; Haney & Zimbardo, 1998; Mischel, 1968; Ross & Nisbett, 1991; Zimbardo, 2004). The Riverside Situational Q-Sort (RSQ: Wagerman & Funder, in press), uses 81 items that describe potentially psychologically important aspects of situations. The present study examines the validity of the RSQ to predict behavioral and emotional outcomes in naturally occurring contexts. 188 undergraduate participants came to the lab 4 times over the course of 4 weeks and described a situation they recently encountered, rated the situation using the RSQ, described their behavior in the situation, and described how they felt in the situation. The results indicate that the RSQ demonstrates impressive validity in predicting behavior with 1569 out of the 5427 possible correlations exceeding r = .15, p < .001. The RSQ also showed strong validity with affect as 23 out of the 81 RSQ items were correlated with Positive Affect at r >= .15, p < .0001 and 37 items were correlated with General Negative Emotionality at r >= .15, p < .0001. Additionally, personality predicted behavior with 715 out of 6700 possible correlations exceeding r = .15, p < .001 and situations people reported being in with 511 out of the 8100 possible correlations exceeding r = .15, p < .001, supporting a triadic model of relationships between persons, situations, and behaviors (Funder, 2006).

47. Savoring as a Mediator of the Influence of Type A Behavior on Enjoyment

Jennifer L. Smith and Dr. Fred B. Bryant, Loyola University

Although theorists have argued that Type A behavior (TAB) damages one’s capacity to enjoy life (Friedman & Ulmer, 1984), prior research has consistently found that TAB enhances the subjective well-being of young adults (Bryant & Yarnold, 1990; Strube, 1990). These prior studies examined TAB as a unidimensional construct. Yet, there is strong evidence that TAB is multidimensional and that a three-factor model, consisting of Hard-Driving Competitiveness, Rapid Eating, and Rapid Talking (RT), best fit the Student Jenkins Activity Survey (SJAS; Glass, 1977) data (Bryant & Yarnold, 1989). The present study examined the relationship between the SJAS subscales and undergraduates reported enjoyment of a recent vacation. It was hypothesized that: (a) RT, which reflects speed and impatience in social situations, would predict lower reported levels of enjoyment; (b) RT would predict lower levels of adaptive, and higher levels of maladaptive, savoring responses; and (c) savoring responses would mediate the negative relationship between RT and enjoyment. A sample of 764 undergraduates (613 females, 151 males) completed self-report measures, including the SJAS and the Ways of Savoring Checklist (WOSC; Bryant & Veroff, 2006). The WOSC consists of a series of statements reflecting things that people might think or do while they are going through positive events. Bryant and Veroff (2006) reported evidence supporting the reliability and validity of 10 WOSC subscales reflecting cognitive and behavioral responses to positive events, including Memory Building, Counting Blessings, and Kill-Joy Thinking. Respondents also reported how much they had enjoyed their vacation. Confirming predictions, multiple regression analyses revealed that: (a) when all three SJAS subscales were entered simultaneously, only RT showed a significant negative relationship with enjoyment; (b) RT predicted less Memory Building, less Counting Blessings, and more Kill-Joy Thinking; and (c) both reduced Counting Blessings, and heightened Kill-Joy Thinking, mediated the dampening effect of RT on enjoyment.

48. Social Motivation In Personality Disorders

Christie T. Spence and Thomas F. Oltmanns, Washington University in St Louis

Personality dysfunction is commonly thought of as being exhibited via difficulty in interpersonal relationships. Many of the Diagnostic and Statistical Manual of Mental disorders (DSM-IV) criteria for the specific Personality Disorders (PDs) speak to this issue directly: persistently bears grudges (Paranoid PD); failure to conform to social norms (Antisocial PD); difficulty making everyday decisions without excessive advice from others (dependent personality disorder); and a pattern of intense interpersonal relationships characterized by alternating between extremes of idealization and devaluation (Borderline PD) (APA, 2000). Many of these difficulties may reflect issues that involve social motivation, i.e. a person's desires and goals (Oltmanns, 2007). The impact of this dysfunction is not measured by the motives themselves, but by the problems with others that may result from them. Two important motives or goals in understanding personality are agency and communion (also known as power and love or achievement and affiliation). Agency, in its most extreme form, represents a desire for dominance, achievement and power. It is often characterized by a focus on the self and a propensity to separate the self from others. Communion, on the other hand, is characterized by a focus on others and an inclination to merge or unite with others (Helgeson, 1994; Lieblich, Zilber & Tuval-Mashiach, 2008; McAdams, 1996). Certain personality disorder symptoms may reflect maladaptive variations in needs for agency and/or communion. For example, one of the criteria for Obsessive-compulsive PD is excessive devotion to work to the exclusion of leisure activities. This criterion represents an extreme form of agency. In this study, we utilize an adaptation of the Life Narrative Interview to probe for motives of agency and communion in an effort to understand the relationship of these motives to Personality Disorders.

49. When do personality traits predict personal goals?

Nick Stauner, Tierra S. Stimson, Michael Boudreaux, and Daniel J. Ozer, University of California, Riverside

While there is evidence that motivational units, from specific personal goals to broad motive dispositions, are related to personality traits, these relations are insufficient to provide clear, comprehensive linkages between motive and trait domains. In this research, we examine the hypothesis that goals that pertain directly to personal characteristics that might be altered are those most likely associated with personality traits. Goals that arise from norms, roles, and life circumstances are less likely to be associated with personality traits. A total of 690 undergraduate participants completed a questionnaire packet that included a measure of the Big Five personality factors (John, Donahue, & Kentle, 1991) and a goal questionnaire, where participants rated the importance of 65 goals. Twenty goal-item clusters were formed on the basis of item content, and correlations between unit-weighted composites of these goal clusters and the Big Five were examined. As predicted, the largest correlations were obtained with goals pertinent to changing personal attributes. Thus goals related to reducing negative affect (e.g., "stop worrying so much", r= .54) were strongly associated with Neuroticism, and goals referring to reducing social inhibition (e.g., "be less shy", r= -.57) were strongly associated with low Extraversion. In contrast, goals related to academics (e.g., "do well in school") were only weakly related to personality traits. In some college students, Introversion and Neuroticism are associated with self-altering aspirations toward Extraversion and Stability. Other goals, such as those relating to finances, health, and family relations, are not associated with parallel personal characteristics.

50. Reexamination of content validity of ACS-2 (Assumed-Competence Scale 2nd version)

Kuniko Takagi, Seirei Christopher University

Tomomi Niwa, Nagoya University

Hayamizu, Kino, and Takagi (2003) proposed a construct "Assumed competence (AC)" to describe emotional characteristics of Japanese youth. They purported that adolescents with high AC looked competent, but in fact, their confidence was derived from undervaluing others. Based on the essence of AC, Hayamizu, Kino, Takagi, and Tan (2004) constructed ACS-2 (Assumed-Competence Scale, second version) to measure AC. It consists of 11 items that describe the tendency of undervaluing others like "There are a lot of insensitive people around me". The items themselves seem to measure their evaluation on others, and not their own competence. Earlier researches showed that ACS-2 and Self-Esteem (SE) (Rosenberg, 1960) were uncorrelated (ex. r=.08, in Hayamizu, et al., 2004), but this result was insufficient to show the relationship between ACS-2 and actual competence or performance. Therefore the purpose of this study was to confirm the meaning of AC by examining relationships within ACS-2, self-evaluate competence (academic competence and grandiosity) and academic performance (estimate and actual academic performance). In other words, to examine whether people with high AC "they pretend to be competent" or "they have baseless competence".
Undergraduates who took psychology classes were asked to rate the questionnaire which contained ACS-2, SE, academic competence scale, and grandiosity scale. They were also asked to estimate their score after attempting 50 multiple-choice questions on psychology, and 154 male and 146 female were subjects of analysis in this research. The correlation analysis result of ACS-2, SE, academic competence, grandiosity, and estimate and real exam score showed that ACS-2 correlates positively with academic competence and grandiosity but uncorrelated with SE, neither estimate nor real academic performance. That is to say, AC indicated by ACS-2 is a facet of competence, and it doesn't concern with expectation for performance or actual result in academic record.

51. Person-Descriptors Ubiquitous Across Cultures: A study of Twelve Diverse Languages

Amber Gayle Thalmayer and Gerard Saucier, University of Oregon

Tarik Bel-Bahar, Anna Freud Centre, London

The discovery of a structural model of personality attributes with a high degree of generalizability across diverse cultural settings is important for many reasons. In an increasingly global communication network, we need a system for communicating about personality characteristics that is not strongly biased toward the concepts and conceptual organization from one culture or nation. Not only is the standardization of measures on Western samples prior to cross-cultural tests a covert form of cultural suppression, but a widely generalizable structural model would better retain its validity and relevance in a wider range of cultural settings and would thus allow for more consistent replicability. The identification of between-culture commonalities would also contribute to the understanding of between-culture differences. To this end, twelve languages representing diverse cultural characteristics and language families, from multiple continents, that are isolated from one another were chosen. Important also was the existence of an extensive dictionary providing English translations of indigenous terms for each language. Diversity of cultural characteristics in a way approaching representativeness of human cultures worldwide was indexed by a major ethnographic atlas. The languages included are (from Africa) Maasai, Supyire Senufo, Khoekhoe, Afar, (from Asia) Mara Chin, Hmong, (from Australia, New Guinea, and Melanesia) Wikmungkan, Enga, Fijian, (from the Americas) Eastern Canadian Inuktitut, Hopi, and Kuna. Every person-descriptive term in each of these dictionaries was cataloged, and a composite list was examined to determine the dispositional content (in terms of English translation) most ubiquitous across languages. At least 20 single word concepts used to describe persons could be identified in all of the study languages. Findings indicate that some personality-attribute concepts may be as cross-culturally ubiquitous as “basic emotion” concepts.

52. Personality Comes out of the Closet: The Unexpected Emergence of "Personality" in an Analysis of Article Titles from the Journal of Research in Personality, 1973-2008.

Gregory D. Webster, University of Florida

With the advent of the Association for Research in Personality's (ARP) first stand-alone conference, it's time to take stock of the topics studied by personality psychologists and how they have changed over time. To this end, I examined words from the titles of every article published in ARP's official journal, the Journal of Research in Personality, from its inaugural issue in 1973 through 2008 (N = 1,465 articles, excluding editorials and comments). For analyses, the 36 years of data were grouped into 4 bins or time windows of 9 years each. Between the periods of 1973-1981 and 2000-2008, instances of "Personality" increased from appearing in 6.7% of titles to 32.2% (+380.6%), perhaps signaling personality psychology's arrival and acceptance as a viable and visible science. Implying increased integration between social and personality psychology, "Social" grew from 4.0% to 6.9% (+72.5%). Personality psychologists also appeared to embrace the scientific study of the self, as "Self-Esteem" increased from 1.6% to 4.9% (+206.25%). A greater emphasis on modeling may have also occurred, because "Model" grew from 1.1% to 5.5% (+400.0%). Perhaps echoing psychology's shift away from studying actual behavior (see Baumeister, Vohs, & Funder, 2007), "Behavior" decreased from 7.0% to 4.9% (-30.0%). Additionally, "Aggression" abated from 8.0% to 2.8% (-65.0%), reflecting aggression research's popularity in the 1970s and subsequent decline. During 1973-1981 (n = 374), the top 5 words were "Effects" (16.8%), "Aggression" (8.0%), "Control" (7.5%), "Behavior" (7.5%), and "Personality" (6.7%). During 1982-1990 (n = 327), the top 5 words were "Effects" (11.0%), "Anxiety" (7.3%), "Behavior" (7.0%), "Control" (6.7%), and "Personality" (6.7%). During 1991-1999 (n =270), the top 5 words were "Personality" (17.4%), "Social" (8.1%), "Effects" (7.4%), "Model" (6.3%), and "Self-Esteem" (5.2%). During 2000-2008 (n = 494), the top 5 words were "Personality" (32.2%), "Differences" (7.1%), "Social" (6.9%), "Individual" (6.1%), and "Model" (5.5%).

53. SNAP Trait Profiles as Valid Indicators of Personality Pathology in Non-Clinical Samples

Joshua Wilt, Benjamin Schalet, C. Emily Durbin, Northwestern University

The validity of the categorical model of personality disorders (PDs) has been called into question by high levels of comorbidity between categories and high levels of symptom heterogeneity within categories. These limitations have led to the rise of several dimensional models of PDs that address the limitations of the categorical model and provide the added advantages greater diagnostic flexibility and increased descriptive information. Among dimensional models, Clark's (1993) model may be particularly useful for its ability to predict dysfunction and distinguish between different PDs in clinical samples. The purpose of the present study is twofold: 1) to determine whether trait profiles constructed from traits in Clark's (1993) model are valid indicators of different personality pathologies in non-clinical samples; and 2) to determine whether the structure of normal personality differs between groups with different levels of maladaptive trait profiles. Two independent samples from a university and a community setting completed the Schedule for Nonadaptive and Adaptive Personality - Short Report Form (Harlan & Clark, 1999) paragraph descriptor assessment of traits included in Clark's model of personality pathology. Self- reports of normal personality traits and of Axis I pathology were obtained for both samples, and informant reports were also collected for the university sample. In reference to our first goal, SNAP trait profiles representing key features of Borderline PD (BPD), Schizotypal PD (SPD), Avoidant PD (APD), and Obsessive-Compulsive PD (OCPD) were modestly intercorrelated and differentially related to measures of normal personality and psychopathology. Importantly, PD profiles predictably related to self-and informant-reports of traits that are considered most relevant to each respective personality pathology (e.g., BPD was strongly related to neuroticism). In reference to our second goal, correlation matrices and cluster solutions of MPQ (Tellegen, 1982) traits differed between groups of individuals with high and low levels of each PD profile.

54. Ostracism and Aggression: The Moderating Influence of Psychopathic Traits

James H. Wirth, Donald R. Lynam, and Kipling D. Williams, Purdue University

Ostracism, being excluded and ignored, thwarts basic fundamental needs and can lead to aggression (Williams, 2007). Psychopathy is characterized by shallow affect, impulsivity, lack of remorse, and disregard for others (Cleckley, 1941; Hare, 1991). We investigated if the psychopath will be immune to the effects of ostracism, or will the psychopath become incensed (susceptible) and more likely to aggress? This may depend on the stage of ostracism. Seventy-two participants completed a brief personality measure (Five Factor Model; McCrae & Costa, 1990). Participants then were randomly assigned to be ostracized or included during an online ball-tossing game, Cyberball. Immediately after Cyberball (reflexive stage) participants completed an assessment of their basic needs and mood, then again after a one minute delay (reflective stage). Participants indicated how tempted they were to act aggressively towards the Cyberball players. Psychopathy moderated ostracism’s reflexive effect on positive mood, B=.86, p=.06, but not negative mood or basic needs. In the reflective stage, psychopathy moderate ostracism’s impact on basic need satisfaction, B=.70, p=.05, negative mood, B=-.96, p=.07, and aggressive behavior temptations, B=-2.48, p<.005, but not positive mood. For basic need satisfaction, the exclusionary status effect was less for those high in psychopathy, B=-.76, p<.001, than those low in psychopathy, B=-1.34, p<.001. For those low in psychopathy there was an exclusionary status effect for negative mood, B=.97, p<.01, and aggressive behavior temptations, B=2.78, p<.001. However, for those high in psychopathy there was no exclusionary status effect for negative mood, B=.17, p=.57, or aggressive behavior temptations, B=.70, p=.13.
Individuals high in psychopathy are susceptible to ostracism’s immediate impact. However, psychopathy does moderate persistent effects of ostracism. Ostracized individuals high in psychopathy felt no different on negative mood and temptations to act aggressively than their included counterparts. It appears those high in psychopathy were not incensed and therefore were not aggressive.

55. The Personality Traits of Liked People

Jessica Wortman and Dustin Wood, Wake Forest University

There has been surprisingly little research on how personality traits are associated with being generally liked by others. Much of the work done has involved identifying the personality traits of popular children, and has found high extraversion and agreeableness are associated with greater peer acceptance (e.g., Jensen-Campbell et al., 2002). The current study had two aims: (1) to document the personality traits that are associated with being generally liked by peers in an older population (college students), and (2) to identify the narrower aspects of broad traits such as extraversion that are most associated with general likeability. Using two samples, we examined the relation between personality traits and liking over several months. First, a sample of students living in dormitories (N≥139) rated their own personalities and their liking of individuals on their hall in two consecutive semesters. Second, a sample of members from seven fraternities and sororities (N≥262) rated their own personalities and their liking of other individuals in their organizations two times a year apart. We found that individuals high on sociability aspects of extraversion (tendencies to be happy, good-humored, cheerful) tended to be more liked, but people high on dominance aspects of extraversion (tendencies to be bold, assertive, forceful) were significantly less liked. Additionally, individuals high on aspects of agreeableness (tendencies to be courteous, kind-hearted, warm) were significantly more liked by peers, and individuals high on aspects of neuroticism (tendencies to be more moody, temperamental, unstable, and less calm, level-headed) were significantly less liked. The results of this study give a detailed personality profile of a person who is generally likeable, as well as demonstrating that the personality traits associated with being generally liked may be fairly similar in different settings.

56. Racial identity and life satisfaction among a community sample of African American men and women.

Stevie C.Y. Yap, Isis H. Settles, and Jennifer S. Pratt-Hyatt, Michigan State University

Racial identity has been conceptualized as a stable, multi-dimensional construct. Racial centrality, private regard and public regard are three facets of racial identity which that may have important implications for well-being among African Americans. There is at least some existing support for each of these dimensions being associated with psychological outcomes for African-Americans. To further understand these complex relationships, we propose that associations between the racial identity dimensions and well-being may be mediated by different identity functions, or perceptions of positive and negative aspects of identity. In the present study of a community sample of 161 African American men and women, we examined whether the relationship between life satisfaction and three racial identity dimensions (centrality, private regard and public regard) were mediated by three identity functions: support and belongingness to family and group, perceptions of discrimination, and resilience. We tested a separate model for each identity dimension in which the three identity functions were simultaneously tested as the mediators. Our results indicated that racial centrality, public regard, and private regard were positively associated with life satisfaction. Support and belongingness to family and group mediated two relationships: racial centrality and life satisfaction, and private regard and life satisfaction. Moreover, the relationship between public regard and life satisfaction was mediated by perceptions of discrimination towards one’s racial group. Implications of these findings on social identity theory and the functions of racial identity are discussed.

57. Attachment Style and Perception of Facial Expressions of Emotion among Close Friend Dyads and Casual Acquaintance Dyads

Fang Zhang and Maria Parmely, Assumption College, Worcester, MA

It is often assumed that intimacy and familiarity will lead to better and more efficient emotional communication between two individuals. However, research so far has failed to unequivocally support this claim. The present study proposes that efficiency in emotional communication in close dyads resides more in the detection of subtle, temporal, and dynamic facial cues than in the detection of full facial expressions. Furthermore, processing efficiency is influence by attachment style of the individual, independent of his or her big-five personality characteristics. Forty-three close friend dyads were compared with forty-nine casual acquaintance dyads on their recognition of the partner’s partial facial expressions, and their attachment style and big-five personality characteristics were measured. The results show that close friends dyads were more accurate than casual acquaintance dyads in detecting each other’s partially formed sad or angry expressions, but the two groups were similarly accurate in detecting happy expressions. Furthermore, secure attachment was related to greater efficiency and accuracy in processing negative facial expressions, especially among close dyads. The findings have important implications for research on communication of emotions in close relationships, calling attention to the need to conduct analyses at the micro and transitional level.

58. Personality Dynamics and Academic Outcomes in First-Year University Students

Andrew J. Wawrzyniak, University of Edinburgh/University College London

Martha C. Whiteman – University of Edinburgh

Mean-level, rank-order, and individual-level trait dynamics in first-year university students was examined in relationship to academic outcomes. The Big Five Domains were measured using a 100-item IPIP (Goldberg, 1999) online inventory in 187 participants (62 males, mean age = 18.75 years, SD = 0.67; 125 females, mean age= 18.73 years, SD = 0.64) assessed at the beginning of both the first and second semester, final exam time, and the beginning of the second academic year. Year-end exam marks were recorded. Mean-level Conscientiousness significantly increased between the beginning of the second semester and exam time (60.92 vs. 66.31, P < .05, d = .44) along with Openness between the start of the first year and exam time (73.08 vs. 76.08, P < .05, d = .29); no other significant mean-level changes were found. All traits showed high rank-order stability between all assessments (P< .01). However, Neuroticism, Extraversion, and Conscientiousness had individual-level reliable change index score distributions significantly different from a normal distribution between the first two assessments; only individual-level Conscientiousness change differed from chance between the beginning of the academic year and exam time (all χ² significant at the 5% level). The only significant correlation between exam scores and individual-level change was only found for Conscientiousness between the first two assessments (ρ = .26, P < .05) indicating that higher exam marks were associated with more individuals increasing on this trait between the start of their academic careers and after first receiving academic feedback. Conscientiousness at any one time point did not correlate with exam marks. These findings suggest that individual-level trait change occurs during the first year of university with meaningful implications for academic outcomes.

59. Support for a ‘Big Six’ Model of Personality Attributes in Inclusive Lexical Studies

Gerard Saucier, University of Oregon

Although the Big Five is the most commonly used structural model for personality attributes, a six-factor model has appeared at least as robust in lexical studies. Unfortunately, previous evidence for both models has been drawn almost entirely from studies with relatively narrow selections of attributes. A study examined factors from previous lexical studies using a wider selection of attributes, in seven languages (Chinese, English, Filipino, Greek, Hebrew, Spanish, and Turkish), and found six recurrent factors, each with common conceptual content across most of the studies. The previous narrow-band six-factor model outperformed the Big Five in capturing the content of the six recurrent wide-band factors. In studies examining a wider range of attributes, the narrow-band Honesty factor tends to morph into a Negative Valence (versus Propriety) factor, and its Emotionality factor tends to morph into Internalizing Negative Emotionality (versus Resiliency). In an American community sample, adjective markers of the six recurrent wide-band factors showed substantial incremental prediction of important criterion variables, over and above the Big Five. Isomorphism between these wide-band six and the narrow-band six factors indicate they are variants of a ‘Big Six’ model that is more general across variable-selection procedures. The stronger relative cross-language recurrence of the Big Six factors gives somewhat better prospects for factorial invariance across populations than is likely for the Big Five.

60. Exploring the Common Lexicon as a Basis for Structural Personality and Personality Disorder Research

Leonard J. Simms, William R. Calabrese, and Monica Rudick, University at Buffalo, The State University of New York

Structural models of personality and personality disorder (PD) are strongly influenced by variable selection decisions. Lexical, dictionary-based studies laid the foundation on which the Big Five model was built, but some have criticized the Big Five, arguing that the subset of descriptors chosen to derive it was too exclusionary and included many uncommon terms that are misunderstood by participants. The comprehensiveness of the Big Five is very relevant to PD research, as some have suggested that Axis II be replaced with dimensions based on the Big Five. In the present studies, we explored the common lexicon as a basis for structural personality and PD research. Rather than sampling descriptors from the dictionary, we asked 579 undergraduates to generate 50 terms to describe each of 6 targets: self, good friend, enemy, romantic partner, parent, and a person with mental health problems. Results revealed 12,611 unique descriptors, of which relatively few were frequently used. We then constructed a questionnaire comprised of the top 250 descriptors from each target type (CL-519) and administered it to a second sample of 550 undergraduate friendship dyads. Structural analyses of the CL-519 revealed clear evidence of additional broad personality factors beyond the Big Five, including factors tapping Depravity, Maturity, Arrogance, and Dominance, as well as other characteristics more peripherally related to personality. The implications for structural research in basic and applied clinical research settings will be emphasized.

61. Oddity: The Sixth Factor of Personality

Michael Chmielewski, David Watson, Lee Anna Clark, University of Iowa

In the past decade, there has been an explosion of research examining the links between personality and psychopathology. For example, it now well established that normal personality traits, particularly neuroticism, have strong ties to both the Axis I and Axis II disorders. In fact, the relationships are so strong that researchers have begun to create structural schemes incorporating both normal personality and psychopathology into a single unified model. There is also is widespread dissatisfaction with the current categorical system of personality disorders (PDs) and increasing agreement that it should be replaced with a dimensional model. One leading contender is a dimensional model of maladaptive personality traits that roughly corresponds to the Big Five. This model, however, contains only four dimensions, as research has failed to find substantial ties between Openness and psychopathology. Furthermore, some researchers have suggested that the model is incomplete as it fails to include characteristics related to Cluster A, the ‘‘odd or eccentric’’ PDs. We present structural data from a variety of samples that suggest the existence of a 6th factor of personality. This factor, which we refer to as Oddity/Peculiarity, is distinct from Openness and the other Big Five dimensions. This Oddity factor not only contains content from Cluster A but it also includes features associated with OCD and dissociation. This content has been excluded from models of normal personality, in part, because the constructs it subsumes are difficult to access via single words and thus have been largely excluded from lexical analyses. Moreover, words relevant to the oddity domain (e.g. odd, eccentric, peculiar) were considered “social evaluations” and thus were excluded from analyses that eventually lead to the creation of the Big Five. Therefore, we believe that an expanded “Big Six” scheme is necessary for a more comprehensive unified model of normal and abnormal personality.

Poster Session #2. Saturday, July 18, 8:30 – 9:45 AM., Grand Ballroom

62. Mathias Allemand. Long-term correlated change in personality traits: A comparison of middle-aged and older adults.

63. Jim Anderson and James Grice. Cognition and personality in binary choice tasks.

64. Ivana Anusic and Ulrich Schimmack. Halo factor in personality ratings.

65. Ashley Ausikaitis and Allan Clifton. Facebook profiles as measures of personality and self-enhancement.

66. Stefanie Badzinski. The relationship between implicative dilemmas and measures of psychological well-being.

67. Gregory Bartoszek and Daniel Cervone. An implicit measure of discrete emotional states: A preliminary investigation.

68. Sarah C. Bienkowski and Mark C. Bowler. A conditional reasoning measure of goal orientation.

69. Terry K. Borsook. Painkilling effects of social interactions.

70. Erika Brown, Jim Anderson, Stefanie Dorough, and James Grice. The Dynamic Analog Scale: A single-item method for personality measurement.

71. Jasmine Carey and Delroy L. Paulhus. Are free will and determinism incompatible?

72. Patrick J. Carroll and Robert M. Arkin. The desirability of alternative selves as a moderator of disengagement from existing desired selves: Stepping up rather than down in revising desired selves.

73. A. Daniel Catterson, Joshua S. Eng, and Oliver P. John. I think I can . . . I think I can: Self-efficacy and the use of emotion regulation strategies.

74. Chmielewski, M., & Watson, D. Affect, personality, and psychopathology: The long-term stability and predictive validity of trait measures across young adulthood.

75. David C. Cicero and John G. Kerns. Multidimensional factor structure of positive schizotypy.

76. Keith S. Cox. Psychological well-being among slum dwellers, sex workers, and other impoverished adults in Nicaragua.

77. Kirby Deater-Deckard, Charlie Beekman, Stephen A. Petrill, and Lee A. Thompson. Dispositional frustration/anger in childhood: Independent genetic links with fear and approach.

78. Christopher Ditzfeld and Carolin Showers. Self-structure and affect valuation: The preference for low arousal.

79. Nicholas R. Eaton, Robert F. Krueger, Susan C. South, Leonard J. Simms, and Lee Anna Clark. Finite mixture modeling of pathological personality dimensions: Identification and validation of personality disorder prototypes.

80. Michael A. Faber. Personality, media preferences, and current concerns.

81. Patrick Gallagher, Rick Hoyle, and William Fleeson. A multiple-parameter, self-report trait measure: Can people describe their own density distributions?

82. Lindsay T. Graham, Cindy K. Chung, James W. Pennebaker, and Samuel D. Gosling. What’s in a name? Consensus and validity of impressions based on online screen names.

83. Gareth Hagger-Johnson. Conscientiousness and mental health: Education and multiple health behaviours do not explain the association.

84. Kathrin J. Hanek, Brad Olson, and Dan P. McAdams. Political orientation, happiness, and the psychology of Christian prayer.

85. Patrick C. L. Heaven and Joseph Ciarrochi. Personality predictors of peer-rated adjustment and likeability: A three-year longitudinal study.

86. Krista Hill and C. Randall Colvin. Positive illusions in romantic relationships.

87. Jacob B. Hirsch, Colin G. De Young, Xiaowen Xu, and Jordan B. Peterson. Bleeding heart liberals and conscientious conservatives: Personality and political ideology.

88. Shannon E. Holleran and Matthias R. Mehl. The accuracy of personality judgments at zero-acquaintance: A meta-analysis using realistic, everyday environments.

89. Lauren J. Human and Jeremy C. Biesanz. The role of adjustment in perceptive and expressive accuracy.

90. Lasse Meinert Jensen. Personal ways of handling everyday life.

91. Yuliya Kotelnikova and Jennifer L. Tackett. Personality correlates of cross-cultural differences in values.

92. Robert D. Latzman, Jatin G. Vaidya, and Lee Anna Clark. Components of disinhibition (vs. constraint) differentially predict aggression and alcohol use.

93. Liat Levontin. Victory with no victims: Amity achievement goals.

94. Michelle Luciano, Jennifer Huffman, Lina Zgaga, Caroline Hayward, Veronique Vitart, Harry Campbell, Alan Wright, Ian Deary, and Igor Rudan. Genome-wide association of personality and psychological distress traits in a Croatian population.

95. Cade D. Mansfield, Kate C. McLean, and Jennifer Pals Lilgendahl. Does wisdom matter in the narrative processing of traumas and transgressions.

96. Kristian E. Markon. Reference reliability: Summarizing measurement precision under conditions of maximal test utility.

97. Ashley E. Mason and David A Sbarra. Thin slices of well-being: Perceptions of psychological adjustment following marital separation.

98. Kira McCabe, Lori Mack, and William Fleeson. Methodology standards for palm pilot experience-sampling studies.

99. Theresa A. Morgan, Michael Chmielewski, and Lee Anna Clark. Relations between trait dependency factors, “depressive” dependency, and normal personality.

100. Kumiko Mukaida, Lauren S. Crane, and Hiroshi Azuma. How people describe their past efforts: A comparison between China, Japan, and the U. S.

101. Kristin Naragon and David Watson. The structure of extraversion and facet-level relations with psychological symptoms.

102. Brady D. Nelson and Stewart A. Shankman. Do individual differences in trait emotional/motivational tendencies predict emotional responses to predictable and unpredictable aversive events?

103. Atsushi Oshio. Aspects of everyday life deemed important by dichotomous thinkers.

104. Sunwoong Park and Jack J. Bauer. Lack of effort, it is my responsibility? It depends on who you are.

105. Carly Parnitzke and Mike Furr. Behavioral manifestation of sub-clinical personality pathology in brief social interactions.

106. Erik Pettersson and Erik Turkheimer. Structural relations between personality and psychopathology free from evaluation.

107. Holly Rau, Paula Williams, Yana Suchy, and Sommer Thorgusen. Openness to experience and efficiency of attentional networks.

108. Katherine Rogers and Dustin Wood. Perceptions of differences in personality traits across U. S. regions are more accurate than chance.

109. J. Philippe Rushton and Paul Irwing. A general factor of personality (GFP) in four personality disorder inventories.

110. Benjamin Schalet, C. Emily Durbin, and Elizabeth Hayden. Hypomanic personality traits: Evidence for unique associations with normal personality dimensions.

111. Leigh Sharma and Lee Anna Clark. The impulsive-like traits.

112. Karen Sixkiller, Grant W. Edmonds, Joshua J. Jackson, Jennifer Fayard, Tim Bogg, Kate E. Walton, Dustin Wood, Peter Harms, Jennifer Lodi-Smith, and Brent W. Roberts. The relationship between lower order structure of conscientiousness and health behaviors.

113. Kathy Smolewska, Jonathan Oakman, and Marta Szepietowska. Emotional experience and borderline personality disorder: Examination of correlates in an undergraduate sample.

114. Juliane M. Stopfer, Mitja D. Back, Simine Vazire, Sam Gaddis, Stefan C. Schmukle, Boris Egloff, and Samuel D. Gosling. Facebook profiles reflect actual personality, not self-idealization.

115. Yusuke Takahashi, S. Yamagata, C. Shikishima, K. Ozaki, K. Nonaka, and J. Ando. Positive parenting received in adolescence does not moderate the genetic and environmental etiology of Big Five personality in early adulthood.

116. Maine Tobari. Prosocial behavior and trait empathy in adolescents.

117. Beth A. Visser, Michael C. Ashton, and Julie A. Pozzebon. Is anxiety part of the psychopathy construct?

118. Sylia Wilson and C. Emily Durbin. Multi-method assessment of normal and pathological personality factors: Convergence and incremental contribution to Axis I disorders.

119. Heike Winterheld. Regulatory focus and social support: A dyadic perspective.

120. Edward A. Witt, M. Brent Donnellan, Robert A. Ackerman, and Rand Congor. Planful competence: A personality trait that even sociologists can love.

121. Qiumei Xu and Marie-Elene Roberge. The effect of leaders’ personality and values on individual and group health.

122. Michelle Yik. What’s interpersonal about the Chinese circumplex model of affect?

123. Axel Zinkernagel, F. Dislich, and M. Schmitt. Leads feedback of automatic behavior to a change of explicit self-knowledge? A study in the domain of disgust sensitivity.

124. Brenda Lee McDaniel. Role models and the moral development of at-risk youth.

125. Randy Colvin. To know or like the self? Contributions of accurate self-knowledge and self-esteem to psychological well-being.

126. Christian Brown, Sarah Berger, and Brenda McDaniel. A comparison of spirituality and religion through moral emotions.

Poster Session #2. Saturday, July 18, 8:30 – 9:45 AM., Grand Ballroom

62. Long-Term Correlated Change in Personality Traits: A Comparison of Middle-Aged and Older Adults

Mathias Allemand, University of Zurich

An important developmental question is whether changes in different personality traits are related over time. This research examined correlated change in personality traits over twelve years by comparing middle-aged and older adults. Data come from the Interdisciplinary Study on Adult Development (ILSE). The sample consisted of 330 middle-aged (42 to 46) and 300 older adults (60 to 64 years). Note that the data on correlated change in the older age cohort have already been reported (Allemand, Zimprich, & Martin, 2008). Personality traits were measured with the NEO-FFI. Correlated change in personality traits was examined utilizing latent change models. The results indicated statistically significant medium effect-sized latent change correlations among personality traits in both age groups, except for Neuroticism in the older age cohort. In addition, changes in personality traits were more strongly interrelated in older adults as compared to middle-aged adults. The results indicate substantive commonality in personality trait change over twelve years in both age groups.

63. Cognition and Personality in Binary Choice Tasks

Jim Anderson and James Grice, Oklahoma State University

The important work of Norbert Schwarz and his colleagues over the past twelve years suggests that, like attitude and market researchers, personality psychologists can benefit greatly from the study of the cognitive processes that underlie self-report questionnaires. Trait theorists, in particular, could benefit from cognitive models that explicitly address how individuals process the items on self-report questionnaires, such as the NEO PI-r or 16-PF. We propose one such model that may be of use to personality psychologists; namely, Vladimir Lefebvre’s algebraic model of self-reflexion (AMS-R). This model is based on assumptions that are most consistent with George Kelly’s Personal Construct Theory; for example, judgment processes can be expressed in a binary, hierarchical form, much like Kelly’s bipolar personal constructs. In this presentation we will explain the model and review a number of studies that support the AMS-R as a valid approach for modeling responses to dichotomous scales. We will also present the results of a simple, yet compelling study that demonstrates a fundamental asymmetry in cognition that is central to the AMS-R model. Specifically, we asked 26 participants (12 males, 14 females) to judge pairs of ambiguous stimuli (pinto beans) as either “good” or “bad.” Based on the model, participants should judge the stimuli as “good” with a proportion equal to .625. The observed mean proportion of “good” judgments (M = .624, SD = .167, CI.95: .56, .69) matched the expected proportion very closely, t(25) = .12, p = .91. This result furthermore replicated a previous study conducted by an independent researcher. The supporting evidence for the AMS-R suggests that is may prove useful to personality researchers for developing a deeper understanding of the congnitive underlying their self-report measures.

64. Halo factor in personality ratings

Ivana Anusic, Michigan State University

Ulrich Schimmack, University of Toronto

A consistent finding in literature is that the Big Five personality dimensions are systematically correlated and that much of these correlations can be accounted for by single higher-order factor (DeYoung, 2006; Musek, 2007; Rushton, Bons, & Hur, 2008). The direction of correlations among the Big Five is evaluatively consistent. Thus, neuroticism loads negatively on the higher-order factor, whereas the loadings of extraversion, agreeableness, openness to experience, and conscientiousness are positive. We call this higher-order factor halo because of its evaluative nature (e.g., Thorndike, 1920). Our research indicates that the halo factor is found in self-reports as well as informant reports (parents, peers, spouses). However, whereas ratings of the Big Five personality dimensions converge across raters, halo does not (average cross-informant halo correlation over four datasets is .07). This lack of convergent validity suggests that halo is not a valid higher-order personality factor. Indeed, we show that halo is partially a rating bias as it is moderately correlated with another validated measure of bias. In addition, halo also contains valid variance that reflects biased perceptions of personality traits and is associated with self-esteem. The most important contribution of this research is that it shows that the single factor of the Big Five is not a true higher-order personality factor and demonstrates how it may be removed and examined separately from valid ratings of the Big Five. It is important to consider and remove halo when examining associations between personality traits and other evaluatively valenced constructs (e.g., life satisfaction) because halo may be at least partially responsible for the observed associations.

65. Facebook Profiles as Measures of Personality and Self-Enhancement

Ashley Ausikaitis and Allan Clifton, Vassar College

Research has shown that self-reports not only reveal information about respondents' personalities, but also shed light on their self-illusions, tendencies to self-enhance, and other personality aspects that they themselves are unable or unwilling to report (e.g., Paulhus & Vazire, 2007). Facebook.com is a popular Online Networking System (ONS) in which individuals construct profiles, which can act as a form of self-report. The purpose of this study was to measure several aspects of personality, particularly self-enhancement, by using Facebook to collect self-, friend- and observer-reported information. Participants (N=83) submitted their profile for analysis. One or more friends of each participant (N=112), randomly selected from these Facebook profiles, were asked to rate individual participants' personality and attractiveness. Independent observers (N=33) then rated all participant profiles on the same aspects in order to discover how opinions based solely on profiles differ from opinions based on real-life acquaintance. Participants also completed a battery of previously validated personality scales to determine the relationship between self-ratings and concrete aspects of Facebook profiles. Results indicate that Facebook profiles do reflect self-reports of personality and trait self-enhancement. A discussion of the effect of personality presentation via ONS on acquaintanceships and friendships in a college setting follows.

66. The Relationship Between Implicative Dilemmas and Measures of Psychological Well-Being

Stefanie Badzinski, Okahoma State University

Studies using George Kelly’s repertory grid method have indicated a relationship between psychological well-being and implicative dilemmas, which are cognitive conflicts in which implicit associations between desired and undesired states prevent persons from actualizing their ideal selves. Two studies utilizing a sentence-completion method of repertory grid elicitation explored the relationship between measures of psychological well-being and the number of implicative dilemmas found in participants’ repertory grids. The sentence-completion task elicited participants’ own personal constructs regarding their body-images, personalities, and beliefs. The first study (N = 101) revealed significant correlations between number of dilemmas and standard measures of self-esteem, depression and anxiety (r’s = .22 to .27, p’s < .05). Number of dilemmas was also shown to account for a significant amount of variance in anxiety beyond that accounted for by actual-ideal self discrepancies alone (p < .05). The second study (N = 147) added measures of psychological well-being that were distinctly consistent with Kelly’s theory and also allowed participants to rate how difficult it would be for them to change their actual selves to match their ideal selves on constructs involved in implicative dilemmas. It was hypothesized that anxiety would be highest for participants with constructs on which change was considered to be “very difficult” and lowest for those with constructs on which change was considered “impossible.” In other words, without the perceived possibility of change, participants were expected to be insulated from the anxiety of self-discrepant, dilemmatic constructs. Zero-order correlations supported these predictions when the analyses were based on the theory-consistent measures of anxiety and guilt (r’s = .17 to .33, p’s < .05). Zero-order correlations based on Spielberger’s Trait Anxiety scores were only partially consistent with the hypotheses. Taken together, these two studies provide potential insight into the personal logic that may underlie individuals’ anxiety experiences.

67. An Implicit Measure of Discrete Emotional States: A Preliminary Investigation

Gregory P. Bartoszek and Daniel Cervone, University of Illinois at Chicago

Although psychologists have devised implicit measures of attitudes and motives (Perugini & Banse, 2007; Wittenbrink & Schwarz, 2007), implicit measures of discrete emotional states are lacking. Emotional states commonly are assessed with explicit measures (e.g., the Profile of Mood States, POMS; McNair, Lorr, & Doppleman, 1971). The need for implicit measures of emotional states is particularly acute in that explicit measures sometimes yield theoretically-unexpected results. For example, appraisal models of emotion (Smith & Lazarus,1990) distinguish sadness and anger on multiple appraisal dimensions. People thus rarely should feel both angry and sad. However, explicit measures of anger and sadness commonly correlate positively (e.g., Norcross, Gaudagnoli. & Prochaska, 1984).

Methods and Results: We administered, to 75 participants, both an explicit emotion measure, the POMS (we examine anger and sadness/depression subscales here) and a novel implicit measure. In the implicit measure, participants judged the emotion displayed in 18 emotionally-neutral faces presented for 100ms each. Participants rated whether each face displayed anger, sadness, guilt, or no emotion. We expected that participants’ current emotional state would bias their interpretation of the faces in an emotionally congruent manner. Measures were completed subsequent to hearing a narrative designed to elicit a negative feeling state. As in prior empirical research, but contrary to theory (Smith & Lazarus, 1990), POMS anger and sadness scores correlated positively, r = .44, p<.001. Participants who reported feeling angry were more likely to report simultaneously feeling sad. However, in our novel implicit measure, anger and sadness correlated negatively; people who judged that faces displayed anger were less likely to judge that they displayed sadness, r = -.308, p < .01. The negative correlation between anger and sadness ratings became stronger when we analyzed the proportion of ratings involving some emotional state (rather than “no emotion”), r = -.647.

68. A Conditional Reasoning Measure of Goal Orientation

Sarah C. Bienkowski and Dr. Mark C. Bowler, East Carolina University

Self-report measures of goal orientation are susceptible to response distortion, which leads to inaccurate assessments of an individual’s goal setting and task choice motivations. Conditional reasoning measures provide an indirect way to assess implicit cognitions associated with personality constructs. In the first research study, a conditional reasoning measure of goal orientation is created. This measure indirectly assesses an individual’s implicit theory of intelligence, which is the foundation of goal orientation theory. Data was collected from 500 students at a mid-sized southeastern university in the United States. Item analyses resulted in the modification of some items. In the second study, a validation of the new measure utilizes a series of anagrams as the criterion measure of task choice and persistence. The results of this validation reveal the impact that an indirect measure has on response distortion. Also, this study displays the effectiveness of conditional reasoning measurement format on surveying implicit cognitions.

69. Painkilling effects of social interactions

Terry K Borsook, University of Toronto

Though long believed to be primarily the outcome of tissue injury and peripheral physiological processes, evidence is rapidly mounting that chronic physical pain may instead be largely perpetuated and/or exacerbated by processes in the brain, processes that are impacted by a host of individual difference factors. Perhaps most intriguing is the finding of an intimate link between social pain (the emotional pain arising from social disconnection) and physical pain. Specifically, social rejection has been found to significantly increase physical pain sensitivity, and fMRI studies have shown that the experience of rejection and of physical pain shares some of the same brain structures. Social injury can thus heighten physical pain but the question remains, might repairing social connection soothe physical pain? To test this hypothesis, we employed a pre-post design in which healthy college students were asked to rate the pain intensity and unpleasantness of painful stimuli before and after engaging in a structured activity with a confederate who was instructed to either be very warm and friendly or slightly standoffish. Potential mediator and moderator variables, including attachment style, self-esteem, and mindfulness were also assessed. Participants who experienced the standoffish confederate reported a drop in pain sensitivity, whereas participants who had the highly positive encounter demonstrated no change in pain. These results suggest that even mild social experiences are sufficient to evoke changes in the pain experience. One plausible explanation for these results is that experimental inductions of connectedness may only lead to changes in pain perception in people who are already relatively high in the need for connection (i.e., lonely). Other possibilities, implications for patients, and a discussion of follow-up research underway will be discussed.

70. The Dynamic Analog Scale: A Single-Item Method for Personality Measurement

Erika Brown, Jim Anderson, Stefanie Dorough, and James Grice, Oklahoma State University

Recent studies have demonstrated the viability of measuring constructs using a single item self-report format. In line with this research, the Dynamic Analog Scale (DAS) is a novel technique for generating single-item measures of personality traits. The DAS is comprised of extensive trait definitions (written by the test constructor) and a continuous analog scale on which the test taker simultaneously rates himself or herself as well as other people. The validity of the DAS was supported in an initial study, and in this second study 116 undergraduate students (46 male, 70 female) completed the DAS based on the Big Five personality traits. Each participant completed the DAS by rating the same people on three different occasions. On the first testing occasion, the participants completed the DAS followed by self-report measures of religiosity, drinking behavior, and general affect. Participants then completed the DAS a second time. Approximately 14 days later the participants again completed the DAS as well as the other measures and a standardized 50-item Big Five questionnaire. The immediate and 2-week test-retest reliability and construct validity of the DAS could therefore be assessed. The results were very good, as the immediate reliability of the DAS ratings for self and others was high (.86), as were the test-retest reliabilities (.79 and .76, DAS grids 1 and 2 vs. 3). With regard to validity the correlations between the DAS and the 50-item Big Five questionnaire were modest (Mdn = .58), but the DAS was found to equally predict religiosity, drinking behavior, and affect when compared to the standardized personality questionnaire (significant r’s ranging from .21 to .58). This study therefore offers additional convincing support for the efficacy of the DAS and the viability of single-item measurement.

71. Are Free Will and Determinism Incompatible?

Jasmine Carey and Delroy L. Paulhus, University of British Columbia

Some psychologists have argued for Incompatibilism, that is, that free will and determinism are polar opposites; Others favor Compatibilism, suggesting that both can be true. Instead of assuming that free will and determinism are polar opposites, we aimed to address the issue by developing separate measures of these two concepts and testing their orthogonality. To this end, we developed a theory-based measure with four subscales. Two of them, Free Will and Scientific Determinism are the focus here. Fate and and Chance subscales represent alternative points of view. The 28 items were adminstered to 255 undergraduates. Results indicated acceptable reliabilities for all scales. Beliefs were strongest for Free Will and weakest for Fate. The subscales were largely orthogonal: Free Will and Scientific Causation showed only a small negative correlation (r = -.13, p < .05, 2-tailed). Correlations with other criteria also supported this independence. Religiosity was associated with Free Will (r = .33, p < .001), but not with Scientific Causation (n.s). Similarly, political conservatism correlated only with Free Will (r = .22, p < .001). This pattern continued with (a) Just World belief, which correlated only with Free Will (r = .38, p < .001), and (b) Right Wing Authoritarianism, which correlated with Free Will (r = .35, p < .001). Interestingly, religiosity and authoritarianism also correlated with beliefs in Fate, which, in principle, seems at odds with free will beliefs. Our data supported the compatibilist position. People who believe in the autonomy of their decisions may also believe that their traits or outcomes are determined by science or fate. Moreover, Free Will beliefs are associated with justification of moral responsibility and authority figures, but Determinist beliefs are not.

72. The Desirability of Alternative Selves as a Moderator of Disengagement from Existing Desired Selves: Stepping up rather than Down in Revising Desired Selves

Patrick J. Carroll, PhD, and Robert M. Arkin, PhD, The Ohio State University

Two studies tested whether the effect of threats on old desired selves depends upon the availability of desirable new selves to reengage with. We predicted that people will abandon existing desired selves even in response to partially as well as fully specified threats when there is a desirable alternative self they can substitute whereas people will hold fast to desired selves even in response to fully specified threats when only undesired alternative selves are available to substitute. Participants (N=152) learned about a new 1 year masters program in business psychology and were encouraged to form a desired self as a business psychologist. Next, an alternative 1-year masters program in human ergonomics was described that had a worse, equivalent, or better job placement record (desirability of alternative) than the MBP program (e.g., starting salary). The advisor then informed participants that their GPA fell short of the minimum GPA for admissions and then used past procedures (Carroll et al., in press) to vary the degree to which they specified the threatening discrepancy (unspecified, partially specified, fully specified). As predicted, participants in the undesired vs. other conditions of the alternative self were less likely to show ultimate declines in commitment to applying for the MBP program and elevations in commitment to apply for the MHE program even in response to fully specified threats. Moreover, those exposed to the desired vs. other conditions of the alternative self were more likely to show ultimate commitment declines to applying for the MBP program and commitment elevations to applying for the alternative MHE program when exposed to partially as well as fully specified threats. These findings suggest that the effect of threats on desired selves is not unconditional but depends on the availability of alternative desired selves.

73. I think I can…I think I can: Self-Efficacy and Use of Emotion Regulation Strategies

A. Daniel Catterson, Joshua S. Eng, and Oliver P. John, California, Berkeley

Individuals can use various strategies to regulate (i.e., manage and control) their emotions. One common strategy, reappraisal, regulates emotions at the cognitive level and changes the meaning and thus the emotional implications of an event. Another strategy is suppression, which regulates emotion at the behavioral-response level and inhibits the overt expression of emotion in the face and body. One limitation of past research is it has examined only the frequency with which individuals use these strategies. Another important variable to consider is the self-efficacy individuals experience regarding their ability to use these emotion regulation strategies. Though Tamir et al. (2007) began to examine the influence of self-efficacy on emotion regulation in general, Bandura (1997) argues self-efficacy is a domain-specific self appraisal. To test whether self-efficacy for reappraisal and suppression are distinct and meaningful constructs, we developed a new measure, parallel to the Emotion Regulation Questionnaire (Gross & John, 2003). Participants rate the degree to which they feel capable using reappraisal and suppression when they want to regulate their emotions. Results of two studies indicate that individual differences in self-efficacy for reappraisal and for suppression (a) are internally consistent constructs (mean alpha = .83), (b) are relatively independent constructs (mean r = .32), (c) predict frequency of use for each regulation strategy (mean r = .39), and (d) predict positive emotional and social well-being outcomes for reappraisal, but negative outcomes for suppression. These findings suggest that self-efficacy beliefs for reappraisal and suppression are important factors to consider in the conceptualization of emotion regulation, and point to several directions for future research.

74. Affect, personality, and psychopathology: The long-term stability and predictive validity of trait measures across young adulthood

Michael Chmielewski and David Watson, The University of Iowa

We report results from the most recent wave of the ongoing Iowa Longitudinal Personality Project (ILPP). ILPP is the first study to assess both the Big Five and trait affect across retest intervals of similar length (participants were assessed at approximately 18, 21, 24, and 27 years of age) in a sample (N = 225) that transitioned through young adulthood. This developmental period, which roughly encompasses ages 18-25 (Arnett, 2000), has recently become the focus of considerable research and is viewed as an important developmental transition because of the significant life changes individuals experience during this timeframe. This study had two basic goals. First, we examine patterns of stability across the 9 years of the ILPP to better understand the processes influencing trait stability and change. Second, there is a large and vibrant literature examining the links between personality and psychopathology. However, the large majority of research has examined these links concurrently and far fewer studies have demonstrated the ability of personality to predict psychopathology at a later date. With this in mind, we included measures of depression and schizotypal personality in our most recent ILPP assessment. Therefore, we will examine the extent to which the Big Five and trait affect from previous assessments (starting when participants were 18 years old) can predict participants’ scores on measures of psychopathology nine years later (when they are 27). Given the substantial life changes that occur during the young adult years, this study provides a unique test of personality’s power to predict subsequent psychopathology.

75. Multidimensional Factor Structure of Positive Schizotypy

David C. Cicero and John G. Kerns, University of Missouri

Schizotypy refers to traits similar to schizophrenia symptoms and is related to cluster A personality disorders. Previous factor analytic studies have found a positive schizotypy factor distinct from a negative factor. However, some evidence suggests that the positive factor may itself be multidimensional, but the factor structure of positive schizotypy is still unclear. The current study provided converging evidence through four different analyses that positive schizotypy is multidimensional. First, a factor model with three positive schizotypy factors (paranoia, referential thinking, and cognitive-perceptual) fit the data better than models with fewer than three factors. Second, a factor model with a second-order (i.e., higher-order) positive schizotypy factor fit the data significantly worse than a factor model without a second-order factor in which first-order factors were allowed to correlate freely, suggesting that the second-order factor does not completely account for relations among the first-order factors. Third, a Schmid-Leiman transformation found that even after accounting for the second-order factor that meaningful variance was attributed to the first-order factors. Finally, the three positive schizotypy factors displayed differential relations with five-factor model personality traits. Overall, results are somewhat consistent with the current DSM-IV classification of personality disorders in that the best fitting model included a separate paranoia factor, which is consistent with Paranoid PD being separate from other Cluster A PDs. However, results are somewhat inconsistent with the DSM-IV in that the cognitive-perceptual and referential thinking factors are not analogous to any DSM PD, and schizotypal PD contains aspects of all three positive factors. Additionally, the pattern of relations among schizotypy factors and FFM personality suggests that paranoid and schizoid, but possibly not schizotypal PD, may be easily incorporated into FFM conceptualizations of personality disorders.

76. Psychological well-being among slum dwellers, sex workers, and other impoverished adults in Nicaragua

Keith S. Cox, Northwestern University

There is a lacuna regarding well-being research in the developing world, especially among the poorest in the developing world. Biswas-Diener and Diener (2002) employed a study design in the slums of Calcutta, India to address this lacuna. They found slightly negative global subjective well-being but slightly positive domain specific subjective well-being in their sample. The current study employs the same paradigm and investigates the subjective well-being of female sex workers, city dump dwellers, urban poor, and rural poor, in Nicaragua, Central America. The current study is conceived of as a replication and extension study of the Biswas-Diener and Diener findings. It is hypothesized that similar well-being results will be obtained. To extend their study, personality and social support measures were included in this study. It is hypothesized that personality and social support will significantly predict variance in subjective well-being. Data analysis indicates that many of the groups have slightly negative global subjective well-being, similar to the Biswas-Diener and Diener findings. Very low global subjective well-being of the female sex worker group is the notable exception to this trend.

77. Dispositional frustration/anger in childhood: Independent genetic links with fear and approach

Kirby Deater-Deckard and Charlie Beekman, Virginia Tech

Stephen A. Petrill, Ohio State University

Lee A. Thompson, Case Western Reserve University

Theories of temperament and behavior propose that dispositional frustration/anger is an important component of both behavioral approach or activation, and behavioral withdrawal or inhibition (Derryberry & Rothbart, 2001*). As part of behavioral approach, frustration/anger can facilitate “offensive” instrumental actions to remove obstructions to potential rewards. As part of behavioral inhibition, frustration/anger can motivate defensive actions to minimize potential exposure to punishments. The aim of the current study was to test the independence of these two systems at the phenotypic and behavioral genetic levels of analysis. Mothers rated their 5-9 year old same-sex twins’ (n = 204 pairs) frustration/anger, fear, and approach/positive anticipation, using the Child Behavior Questionnaire—Short Form (CBQ-SF; Putnam & Rothbart, 2006). All three temperament scores varied widely in the sample. Fear and approach/positive anticipation were independent of each other—consistent with past research and theory demonstrating that behavioral approach and inhibition motivation systems are independent (Gray, 1987). However, fear and approach/positive anticipation both were correlated positively with frustration/anger (r = .21 to .28, p < .05). In regression analyses, fear and approach/positive anticipation each provided unique statistical prediction of frustration/anger, together accounting for 10-14% of the variance in frustration/anger. The very same pattern was found in the underlying genetic variance and covariance between fear, approach/positive anticipation, and anger. There was moderate heritable and nonshared environmental variance in all three variables. In addition, there were significant genetic correlations between fear and frustration/anger, and approach and frustration/anger, and no genetic overlap between fear and approach. The findings highlight the importance of distinguishing the genetic etiology of “offensive” and “defensive” frustration/anger operating within independent behavioral activation/approach and inhibition/avoidance motivational systems.

78. Self-Structure and Affect Valuation: The Preference for Low Arousal

Christopher Ditzfeld and Carolin Showers, University of Oklahoma

Showers’s (1992, 2002) model of evaluative self-structure suggests an association between compartmentalization (the organization of positive and negative self-attributes into separate self-aspects) and intensified affective experiences. Integration (the organization of positive and negative self-attributes into the same self-aspects) presumably allows for greater affective stability (cf. Zeigler-Hill & Showers, 2007). To test this, the current study’s design incorporates Tsai et al.’s (2006) affect valuation index (AVI) which accounts for two affect dimensions:

valence and arousal (derived from Russell’s affect circumplex). For the AVI, participants not only report their typical (actual) affect, but also their ideal affect (i.e., affect they value experiencing most). Results from actual affect scales indicate that individuals with integrative self-structures experience more low arousal affect than those who are compartmentalized, in analyses that control for high arousal affect; low arousal positive (LAP), β = -.18, p < .03, and low arousal negative (LAN), β = -.15, p < .09. Moreover, an interaction, β = -.21, p < .03, indicates that the LAN effect is most pronounced for individuals with relatively positive self-concepts. These findings suggest that individuals with integrative structures avoid high arousal affect, even when positive, instead preferring the stability from experiencing low arousal affect. In the process of maintaining stability, people with integrated structures appear to be more tolerant of LAN affect than those who are compartmentalized (they avoid negative affect at any cost). In addition, results from ideal affect scales indicate that integration is associated with valuing LAP affect more than with compartmentalization (β = -.19, p < .03), suggesting that people with integrative structures not only experience more low arousal affect, they appear to place higher value in LAP states (contrary to the idea that most individuals strive to experience high arousal positive affect).

79. Finite Mixture Modeling of Pathological Personality Dimensions: Identification and Validation of Personality Disorder Prototypes