Normal and Abnormal Personality Traits:Evidence for Genetic and EnvironmentalRelationships in the Minnesota Study ofTwins Reared ApartKristian E. MarkonRobert F. KruegerThomas J. Bouchard, Jr.Irving I. GottesmanUniversity of Minnesota, Twin Cities CampusABSTRACT Recent studies have demonstrated substantial correlationsbetween normal and abnormal personality traits. Yet little is known abouthow these correlations are mediated genetically and environmentally: Donormal and abnormal personality traits stem from the same underlying genesand environments? We addressed this question using data from 128monozygotic and dizygotic twin pairs in the Minnesota Study of TwinsReared Apart (MISTRA). Additive genetic and nonshared environmentalcorrelations between scales of the Minnesota Multiphasic PersonalityInventory (MMPI)—an index of abnormal personality— and the Multi-dimensional Personality Questionnaire (MPQ) — an index of normalpersonality— were estimated. Results indicated that phenotypic correlationsbetween normal and abnormal personality were mediated by genetic as wellJournal of Personality 70:5, October 2002.Copyright#2002 by Blackwell Publishing, 350 Main Street, Malden, MA 02148,USA, and 208 Cowley Road, Oxford, OX4 1JF, UK.We thank Auke Tellegen for his helpful comments regarding this article. Kristian E.Markon and Robert F. Krueger were supported by National Institute of Mental HealthGrant MH6S137.Correspondence concerning this article may be addressed to Kristian E. Markon orRobert F. Krueger, Department of Psychology, University of Minnesota, Elliott Hall,75 East River Road, Minneapolis, Minnesota 55455. Electronic mail may be sent [email protected] or [email protected] environmental factors, although the magnitude of genetic mediation tendedto be larger overall. Moreover, the patterns of phenotypic, genetic, andenvironmental relationships among the scales were similar, suggesting thatinfluences on normal and abnormal personality act through systems commonto both. It is suggested that future research focus on the neurogeneticsubstrates of these shared systems and how dysfunction in these systemsinfluences development of disordered personality.Recently, renewed attention has been focused on the relationshipbetween normal and abnormal personality functioning. Distinctionsbetween adaptive and pathological personality have increasingly beenquestioned (e.g., Grove & Tellegen, 1991; Strack & Lorr, 1997), withmany reevaluating how normal personality relates to personalitydisorder and other behavioral pathology (e.g., Krueger et al., 1996;Trull & Sher, 1994). The purpose of the present study was to estimatethe relative contributions of genetic and environmental influences tocovariation between normal and abnormal personality traits, using asample of twins reared apart. By partitioning correlations among acomprehensive set of normal and abnormal traits into genetic andenvironmental components, we sought to specify not only how the twoforms of personality are related, but also why.Existing Models of the Relationship BetweenNormal and Abnormal PersonalityIn part, a renewed focus on relationships between normal andabnormal personality resulted from the introduction in DSM III(American Psychiatric Association, 1980) of the Axis II personalitydisorders (Strack & Lorr, 1994, 1997). The introduction of Axis IIinitiated a number of debates regarding classification and diagnosticstructure of personality disorders, focusing not only on the use ofcategorical versus dimensional schemes, but also on the structurethose categories or dimensions might take (e.g., Grove & Tellegen,1991; Millon & Klerman, 1986). Although these issues have not beencompletely resolved (e.g., Clark, 1999; Loranger, 1999), there isincreasing agreement that abnormal personality is often bettermodeled as being continuously, rather than categorically, distributed(e.g., Livesley, 1991; Widiger, 1993; Widiger & Frances, 1994), andthat the structure of abnormal personality parallels that of normal662 Markon et al.personality (e.g., Costa & Widiger, 1994; Eysenck, 1994; Schroeder,Wormworth, & Livesley, 1992). Numerous studies have demon-strated correlations between continuous measures of normal person-ality traits and personality disorder (e.g., Cloninger & Svrakic, 1994;Costa & McCrae, 1990), as well as joint factor loadings of measuresof normal and abnormal personality (DiLalla, Gottesman, Carey, &Vogler, 1993; Schroeder et al., 1994). These relationships are notlimited to a few particular domains, but rather have beendemonstrated across a number of traits (e.g., Costa & Widiger,1994) and trait models (e.g., DiLalla, Gottesman, Carey, & Vogler,1993; Schroeder et al., 1992).Consensus that normal and abnormal personality are related,however, has not been associated with a similar consensus as towhy this is the case (e.g., Widiger, Verheul, & van den Brink, 1999). Anumber of models have been proposed to explain the relationshipbetween normal and abnormal personality, including hypotheses thatnormal-range personality traits can influence or cause disorder; thatabnormal personality traits can affect other, normal-range, personalitytraits; and that normality and disorder exist in a spectrum, eachrepresenting portions of a continuum (Widiger et al., 1999). There isempirical support for many of these models, and any one or all of themmay account for observed correlations between normal and abnormalpersonality traits to some extent (Widiger et al., 1999).Models such as these are phenotypic in nature. That is, a givenabnormal or normal phenotype is invoked as an explanation for theother, or the two phenotypes are spoken of as existing on a continuum.The emphasis in existing models is on observed variation in normal andabnormal traits, rather than the sources of observed covariationbetween them. An alternative possibility, therefore, is to model therelationship between normal and abnormal traits in terms of theiretiologic antecedents—to ask why, in addition to how, the two types oftraits are related. Especially important in this regard is how covariationbetween normal and abnormal personality traits is genetically andenvironmentally mediated. Normal and abnormal personality traitsmay be correlated, for instance, because they are both directly orindirectly influenced by some common set of genes, a common set ofenvironmental factors, or some combination thereof.One might hypothesize, for example, that environmental