84OLSON VERNON HARRIS AND JANtude o trait ar du t heredit versu th env - PDF document

84OLSON VERNON HARRIS AND JANtude o trait ar du t heredit versu th environmen inonsensical jus lik askin whether a leak basemen i cause b th crac i th foundatio o th wate outside I a ver sense geneti effect ar als environmenta becaus the i a environment an environmenta effect ar alsgeneti becaus the ar mediate b biologica processesNotwithstandin th intertwinin o heredit an environmenti i possibl t estimat th exten t whic difference betwee o a attitud o trai ca b attribute t geneti versuenvironmenta factors Tha is withi a particula rang o envi an geneti "values o "scores (namely th rang ith sample) i i possibl t estimat whether interindividua reflec variatio i th environmenta score and/o thgeneti scores Estimate o geneti influenc ar labele herita- coefficients.Not tha heritabilit coefficient d no indicat th exten twhic characteristic ar geneticall cause (which a mentione i impossibl t determine) Rather the indicat th extent whic variatio o th characteristi i th sampl i attributabl geneti differences An geneti o environmenta influence appl uniforml t al member o th sampl wil no b i thi statistic Fo example "two-leggednes i humani highl geneticall determined but difference i leggednesamon human ar mostl du t environmenta accident (Loeh & Nichols 1976 p 1) thus individua difference i legged woul generat a lo heritabilit coefficien despit thpredominantl geneti basi o two-leggednessHeritabilit coefficient ar dependen o th variabilit oenvironmenta an geneti score i th sample I member o asampl wer geneticall identical the an variatio i phenotyp necessaril b du t environmenta factors I member o sampl ha experience identica environments the an variatio i phenotyp woul necessaril b du t geneti factors Tth exten tha a sampl i representativ o a large populatio i o th heterogeneit o gene an environments heritabilit reflec th percentag o th variatio i a character amon member o th populatio tha i attributabl tgeneti factorsGeneti influence als d no impl tha th characteristi iunchangeable T tak a simpl example hai colo i geneticalldetermine bu i completel malleable Thus learnin tha atti o trait hav significan heritabilitie doe no mea the ar o impossibl t alte (althoug Tesser 1993 ha speculate tha heritabl attitude ma hav a biologica substrat thamake attitud chang mor difficult) Fo instance i member o populatio hav bee expose t ver simila environment wit t a particula characteristi (e.g. the hav al bee ex t extensiv propagand regardin a particula belief) th (belief wil yiel a hig heritabilit coefficien i sampl eve thoug a differen environmen woul havproduce ver differen score (beliefs)Turnin t a mor specifi issue ho coul gene producparticula evaluation o object b individuals First i i extremel unlikel tha ther ar direct one-to-on connection betwee gene an attitude (e.g. a gen tha cause attitude towar punishment o eve many-to-on connection (e.g. a se o that together caus attitude towar capita punishment) gene probabl establis genera predisposition o natura whic the shap environmenta experience i waytha increas th likelihoo o th individua developin specifitrait an attitudes Fo example childre wh ar naturall smal thei ag migh b picke o b othe childre mor tha theilarge peers A a result thes childre migh develo anxietieabou socia interaction wit consequence fo thei personalit(e.g. introversion an thei attitude (e.g. dislik fo parties)Wha ar som specifi geneticall influence characteristictha migh systematicall bia environmenta experience s a tinduc particula attitudes Tesse (1993 identifie severa possibilities includin intelligence sensor structures an temperament Intelligenc i widely—thoug no universally—viewe a a heritabl componen (e.g. DeFries Plomin & LaBuda1987 Snyderma & Rothman 1987) Intelligenc coul plausiblb linke t man attitudes rangin fro enjoymen o specifiintellectua activitie lik readin an ches t attitude towar globa target lik educatio an th rol o computer i oulives I addition i attitude towar controversia issue ar influence systematicall b cognitiv reasoning the intelligenc als contribut t a geneti componen i thos attitudesSensor structure ar inarguabl geneticall based a leas i part coul influenc preference fo variou type o sensor experiences Fo example tast difference coul affec attitude chocolate hearin sensitivit coul affec attitude towarlou music an difference i inner-ea structure tha maintaibalanc coul affec attitude towar rolle coaste rides Personalit an temperamen characteristic hav bee a centra focu obehaviora genetic research Evidenc exist t suppor genetian biochemica component o severa personalit traits includ sociability neuroticism an sensatio seekin (e.g. Bouchar McGue 1990 Loehlin 1989 Loehli & Nichols

1976 Tellege e al 1988 Zuckerman 1995) Thes broa trait coulinfluenc a wid rang o attitudes wit som obviou examplebein sociabilit affectin attitude towar meetin ne peopleneuroticis affectin attitude towar alcohol an sensatio seekin affectin attitude towar skydiving I a nonattitudina domain Jockin McGue an Lykke (1996 foun tha geneti affectin personalit accounte fo a substantia portio oth heritabilit o a differen construct divorc risk Tha is thheritabilit o divorc ris wa show t b relate t geneticallbase difference i personalityThus ther ar man characteristics themselve havin a geneti component tha coul affec individuals attitudes Thes constitut potentia mediator o th genes-attitude relation I othe words thes variable migh b mediatin mech- tha underli connection betwee gene an attitudesPas Researc o Attitud HeritabilitA mentione above researcher hav show tha individuadifference o a variet o broa personalit factor ar partl t geneti difference (e.g. Loehli & Nichols 1976 e al. 1988) Thes factors suc a sociability domi an self-acceptance ar generall presume t encompasman specifi attitudes Ther hav als bee demonstration thaspecifi behavior (whic ma impl specifi attitudes hav significan heritabilities includin alcoho consumptio (Partanen,Bruun & Markkanen 1966 an televisio viewin (Plomin Corley DeFries & Fulker 1990) HERITABILIT O ATTITUDE84Bu wha abou th heritabilit o specifi attitude themselvesA smal numbe o researcher hav examine directl whetheevaluation o individua target (objects behaviors issues etc. a heritabl componentFo example Perr (1973 examine a sampl o 8 pair otwin (approximatel hal identica an hal fraternal) wh com Likert-typ item assessin attitude towar alcohol cigarettes an coffee Previou researcher ha foun tha behavioravariabilit i eac o thes domain (e.g. amoun consumed wapartl attributabl t geneti influences s Perr wante t se th heritabilit o attitude parallele th behaviora findings I fact attitude towar alcoho manifeste a significanheritabilit coefficient wit geneti factor accountin fo 51 oth varianc i alcoho attitudes wherea attitude towar ciga an coffe di no yiel significan heritabilit estimatesScar an Weinber (1981 administere th Authoritarianismscal (Californi F-Scale Adorno Frenkel-Brunswik Levinson &Sanford 1950 t th member o 12 biologica an 11 adoptiv involvin approximatel 45 children Thi scal consiste o 2 sociopolitica statements t whic respondent indicate thei agreemen o disagreement Th result showe generall nonsignifican correlation betwee th tota authoritarianis o member o adoptiv familie (excep husband-wife but significan correlation betwee th tota score o o biologica families O th basi o som furthe involvin I scores th author hypothesized tha thgeneti transmissio o verba abilit partl accounte fo thbiologica componen o authoritarianismArvey Bouchard Segal an Abraha (1989 studie 3 pairo identica twin wh wer reare apart Thes participant com a 20-ite jo satisfactio questionnaire Result showe tota jo satisfactio score yielde a significan heritabilit wit approximatel 30 o th varianc i jo satisfactio bein attributabl t geneti factorsI a relate stud fro th sam laboratory Keller BouchardArvey Segal, an Dawi (1992 administere a wor value inventor t 4 pair o twin wh wer reare apar (2 identicaan 2 fraternal) Si composit score fo wor value wer fro respondents expresse preference fo 2 individua jo outcome (e.g. a composit scor fo th valu attache t"achievemen outcomes") an al si yielde significan heritabil coefficients wit th proportion o varianc accounte forangin fro 18 t 56 (M = 38%)Waller Kojetin Bouchard Lykken an Tellege (1990 surveye 8 pair o twin wh wer raise apar (5 identica an 3 an mor tha 80 pair o twin wh wer reare (approximatel 45 identica an 35 fraternal) Fiv measurin religiou attitude an interest wer b a leas som participants interes i choosin areligiou occupation involvemen i religiou activitie durinleisur time religiou fundamentalism religiou interests anreligiou values Al fiv measure yielde significan heritabilitcoefficients accountin fo betwee 41 an 52 o th interindividua varianceI th larges studie o geneti variabilit i specifi attitudesEave e al (1989 reporte th result o tw surveys on involv almos 3,00 pair o same-se twin i Australi (approxi 2,00 identica an 1,00 fraternal an th othe involvin 70 same-se twin i Grea Britai (approximatel45 identica an 25 fraternal) Australia respondent completea 50-ite measur o conservatis (Wilso & Patterson 1968)whic consiste o one o two-wor item (e.g. "deat penalty,"hippies, "whit superiority, "jazz, an "bibl truth" t whicpeopl answere "yes, "?, o "no. Thirty-thre o th 5 item significan heritabilit estimates whic range fro 21 51 (se Marti e al. 1986) Brit

is respondent complete a publi opinio inventor (Eysenck 1951) whic consiste o attitud statement o a variet o issues includin crimereligion an rac (e.g. "Crime o violenc shoul b punishe bflogging, " perso shoul b fre t tak hi ow life i h t d so withou an interferenc fro society, "Th ma ca liv a goo enoug lif withou religion, an practica ma i o mor us t societ tha th thinker")respondent indicate th exten o thei agreemen o disagreemen wit eac statemen o a 5-poin scale Heritabilit estimate individua item range fro 1 t 62% wit a media 39%Whe th finding acros differen studie ar examined therappear t hav bee som stabilit i th kind o attitude thahav yielde hig heritabilities Thoug no perfect ther ha beea consisten tendenc fo attitude towar th treatmen o crimi (e.g. deat penalty flogging t produc hig heritabilitcoefficients Anothe domai o attitude tha ha appeare heri i religiou attitudes—no specifi religiou affiliations bugenera view abou th valu o religio (se Eave e al. 1989)Attitude towar authorit an towar unconventiona behavio als yielde hig heritabilit estimate i mor tha onstudy I contrast item tha hav yielde littl evidenc o heritabilit hav include attitude towar politica ideologie (e.g.socialism capitalism an attitude towar behavior b younpeopl (e.g. teenag drivers pajam parties)Th Presen StudI th presen study w examine th geneti basi o individuadifference i attitude b collectin dat fro a larg numbe omonozygoti (identical an dizygoti (fraternal twins W assesse a wid rang o attitud topics includin specifi object sweets rolle coaste rides) socia issue (e.g. capitalism o demand) an behavior (e.g. playin chess exercising) Man o thes attitude ha neve befor bee measure itwi research B comparin th correlation withi identica an twins w coul estimat th amount o varianc betwee i attitudina response tha wer attributabl t geneti W als calculate estimate o environmenta influence th attitudesW ha severa goal i thi research On goa wa t tes threplicabilit o pas finding o th heritabilit o attitudes Iindividua difference i som attitude reall ar strongl genet determined the the shoul yiel hig heritabilit coeffi acros studie (although, o course samplin variabilit anmeasuremen erro ca produc fluctuatin results) A mentione i i possibl t identif attitud domain tha hav produce eithe hig o lo heritabilities w wante t confir thes i anothe large-sampl studyA secon (an mor interesting goa was t explor possiblmediator o geneti effect o attitudes W hope t identif tha migh underli gene-attitud relations Thi issu i 84OLSON VERNON HARRIS AND JANcomplex s w wante t begi explorin i b simpl obtaininsom preliminar evidence W measure severa possibl mediator directl throug self-repor assessment o physica charac (athleticism appearance) academi achievemen (a a fo intellectua abilities) an personalit factors W testewhether thes factor wer themselve heritabl and i so whetherthe share geneti varianc wit heritabl attitudes Tha is wexamine whethe attitude an potentia mediator wer correlate a th geneti level Significan geneti correlation ar wit (thoug no definitiv about th hypothesi thaon variable' heritabilit i attributabl t th heritabilit o th variable Fo example i attitude towar readin prove t heritabl an als correlate a th geneti leve wit a measur intelligenc tha wa itsel heritable on interpretatio coul btha attitude towar readin wer t som exten heritabl becaus wa heritable Fo an attitude-mediato pairing thashare geneti variance w als examine mediatio a th phenotypi leve (i.e. whethe th phenotypi covariatio betwee thvariable wa consisten wit a causa mode goin fro th t th attitude) Finally w als conducte analyse thaexplore whethe potentia mediator wer involve i th heritabilit o al o th attitudes no jus particula attitudes Specifically w teste whethe th strengt o th correlation betwee possibl mediator an attitud score predicte th heritabil o th attitudes I attitude tha correlat highl wit amediato variabl yiel highe heritabilitie tha d attitude thad no correlat wit th mediato variable the th mediato b a mechanis throug whic gene affec th se oattitudes Fo example i th heritabilit coefficient o a se o correlate wit th correlation betwee th attitude anth personalit trai o sociability an i sociabilit wer itselheritable the sociabilit woul b implicate a a possibl explanatio o wh o ho th attitude yielde significanA thir goa o ou researc wa t gathe dat relevan t aide put forwar b Tesse (1993) Tesse hypothesize tha atti tha ar highl heritabl migh hav a biologica basi tha attitud chang difficult Becaus peopl fin i difficul ouncomfortabl t chang thes attitude (an therefor d no likt d so) the ma develo psychologica defense t "protectthem A a result thes attitude ma becom

"stronger i a sense Fo example "niche-building migh occu(se Plomin DeFries & Loehlin 1977 Scar & McCartney1983) suc tha individual see ou environment tha ar compatible wit thei highl heritabl attitude o eve construc theiow environment t protec thes attitudesTesse (1993 Creli & Tesser 1996 Tesse & Crelia 1994 hateste thi ide i severa ingeniou ways I al o hi studiesattitude wer selecte fo us tha ha bee show b Eave e al(1989 t hav eithe hig o lo heritabilit coefficients I on respondent wer show t provid thei answer morquickl fo hig tha fo lo heritabilit attitudes I anothestudy individual wer foun t b les affecte b conformitpressur whe reportin hig tha whe reportin lo heritabilit I a thir study interpersona similarit o hig heritabilit attitude was show t affec likin mor tha similarit olo heritabilit items Finally i tw studies peopl foun agree feedbac mor reinforcin whe i wa base o highl attitude tha whe i was base o les heritabl attitudes Al o thes result sugges tha attitud strengt i relate tattitud heritabilityW examine thi issu i a differen way W measure selfreporte attitud strengt fo ever item W expecte tha attitude tha yielde hig heritabilit coefficient woul als yielhig averag strengt ratingsMethoParticipantsPair o adul twin wer recruite throug advertisement i locanewspaper askin fo twin t participat i psychologica researc iretur fo monetar paymen (th specifi amoun pai t participantdepende o th se o questionnaire the wer aske t complete) Twosample o twin wer collected on i London Ontario Canada an on Vancouver Britis Columbia Canada Participant wer involve ieithe th Wester Ontari Twi Projec o th Universit o Britis Twi Project whic ar ongoin behaviora geneti studie o an menta abilityTh ful se o dependen measure reporte i thi articl wa completeb a tota o 67 participants consistin o 19 pair o monozygoti (MZ (12 femal pair an 6 mal pairs an 14 pair o same-se (DZ twin (9 femal pair an 4 mal pairs) Zygosit wa throug a questionnair designe b Nichol an Bilbr (1966)whic ha a reporte accurac o 93% a compare wit th result o typin (Kasrie & Eaves, 1976) Th mea ag o th participantwa 30.4 year (29.9 fo th identica twins an 31.0 fo th fraternatwins)Dependent MeasuresAlon wit othe questionnaire unrelate t th presen article participant complete a attitud surve a hom an returne i b mail Thsurve assesse participants favorabilit towar 3 target (includincontroversia issues persona activities an socia settings) Th ful se oattitud target i presente i Tabl 1 Eac ite wa phrased "M overalattitud towar [nam o issue is " wit possibl answer ranginfro — (extremely unfavorable) throug 0 (neutral) t 3 (extremelyfavorable). Immediatel afte indicatin thei favorabilit towar th target participant answere tw question tha assesse th strengt o theiattitude "Ho importan i thi attitud t you? wit possibl answerrangin fro 0 (not at all important) t 6 (extremely important), an "Hostrongl d yo hol thi attitude? wit possibl answer rangin fro 0{not at all strongly) t 6 (extremely strongly).Participant als provide rating o themselve o 2 personalit traitseac o whic represente a subscal o th Personalit Researc For(Jackson 1984) Thre adjective describe eac trai (e.g. "humble put sel down, "ambitious competitive achieving") anparticipant rate th self-descriptivenes o th trai o a scal fro 0 (notat all) t 6 {very much). Th ful se o trait (a reflecte b on o ththre adjectives i presente i Tabl 2Participant als rate themselve o fou additiona items tw o whicassesse thei physica capabilities "Ho athleti (naturall goo a phys activities ar you? wit possibl answer rangin fro 0 (not at all t 6 (extremely athletic), an "Ho physicall stron ar you?wit possibl answer rangin fro 0 (not at all strong) t 6 (extremelystrong). Th nex tw item assesse participants physica attractiveness"Ho woul yo rate you ow physica attractiveness? wit possibl rangin fro 0 (not at all attractive) t 6 (extremely attractive), "D othe peopl conside yo t b goo looking? wit possiblanswer rangin fro 0 (not at all good looking) t 6 (extremely goodlooking). HERITABILIT O ATTITUDE84Tabl 1Genetic Analyses of Individual Attitude ItemsAttitudDoin crosswor puzzleDeat penalt fo murdeSweetOpen-doo immigratioDoin athleti activitie euthanasiSmokinBeing th cente o attentioSeparat role fo me an womeEducatio racia discriminatio illegaLou musiGettin alon wel wit othe peoplCapitalisPlayin organize sportBi partiePlayin chesLookin m bes a al timeAbortio o demanPubli speakinPlayin bingWearin clothe tha dra attentio acces t birt controExercisinOrganize religioBein th leade o groupReadin book a punishmen fo se crimeBein assertiv coaste rideCorrelationM.4.3.4.4.4.4.3.2.3.3.5.2.5.4.3.4.3.3.3.4.4.5.2.5D.1.3.2.2.3.1.2-.0.4.1.1.3.

2.1.2.1.0.2.3ModeADAACAAAACACADACAAADACADAACCCAAAD •ADACCBes fittin modeFiX2(3) = 1.47 ns\2(4) = 5.61 ns^(3 = 1.41 nsx2(4) = 2.18 nsX2{4) = 2.18 nsX2(4) = 2.29 nsX2(3) = 2.51 nsx2(4) = 5.52 nsX2(4) = 2.54 ns^(4 = 11.64 p .0X2(3) = 4.71 nsx2(3) = 1.15 «X2(4) = 19.61 p .00X2{4) = 4.67 nsX2(3) = 0.46 nsX2Q) = 2.14 nsX2(4) = 2.76 ns^(3 = 3.13 nsX2(4) = 1.00 nsx2(3) = 1.91 nsX2(4) = 7.07 nsX2(3) = 2.39 nsX2(4) = 5.35 nsx2(4) = 2.77 ns^(4 = 3.17 MX2(3) = 2.13 nsx2{3) = 4.31 «X2(3) = 0.48 nsx2(4) = 4.00 nsX2(4) = 2.82 nsa2.0.4.4.4.3.2.0.0.1.3.0.3.3.1.2.0.2.3.1.0.5Estimatec2.0.0.2.2.4.0.1.0.0.1.3.2.2e2.5.5.5.5.5.4.7.7.6.6.4.7.4.5.6.5.6.6.5.6.7.4d2.4.0.0.0.0.3.0.5.3.0.0Note. Estimate o .0 mea tha th componen wa teste bu di no accoun fo a significan amoun o variance M = monozygoti twins D =dizygoti twins A (a2) = additiv geneti variance E (e2) = nonshare environmenta variance D (d2) = nonadditiv geneti variance C (c2) = shareenvironmenta varianceFinally participant wer aske t repor thei grad poin averag(GPA i thei fina yea o hig schoo (a a percentage an th highesleve o educatio the ha completed Thes academi achievemen measure wer take a gros indicant o intellectua ability.Heritability AnalysesAnalyse wer performe t estimat th amount o varianc i attitudresponse (an i othe dependen measures withi ou sampl tha werattributabl t geneti an environmenta factors W conducte thesanalyse o eac individua ite i th questionnair (attitudes personality athleticism attractiveness an grades) a wel a o composit variable forme o th basi o facto analyse (reporte below)Fo eac dependen measure covarianc matrice wer compute be-twee eac pai o twin fo M an D pair separately wit thcompute progra PRELI (Joresko & Sorbom 1989b) A model-fittinapproach b mean o th compute progra LISRE VI (Joresko &Sorbom 1989a) wa applie t thes covarianc matrice t estimat thproportio o th varianc attributabl t additiv geneti (a) shareenvironmenta (c) an nonshare environmenta (e factor (AC model)Th obtaine maximum-likelihoo paramete estimate o thes component wer square t for a2 (percentag o varianc attributabl tadditiv geneti factors—i.e. geneti "mai effects fro th tw parents)c2 (percentag o varianc attributabl t environmenta factor tha wershare b th twins) an e2 (percentag o varianc attributabl t environmenta factor tha wer uniqu t on o th twins) Fo thos dependen measure i whic th ratio o th M correlatio t th D correlatio wa greate tha 2 i wa possibl t tes fo nonadditiv genetieffect (d)—tha is geneti influence resultin fro a interactio o thtw parents contributions Fo thes measures a mode specifyin additivgeneti variance nonadditiv geneti variance an nonshare environmenta varianc wa teste (AD model) Th obtaine maximum-likelihooparamete estimate o thes component wer square t for estimate oa2, d2 (percentag o varianc attributabl t nonadditiv geneti factors)an e2.Thre reduce model fo eac measur wer als teste tha systematicall remove on componen o variance Th firs remove additivgeneti effects th secon remove share environmenta effect (ononadditiv geneti effects) an th thir remove bot type o effects Asignifican chang i th chi-squar analysi fro th ful mode t thchi-squar analysi fro a reduce mode indicate tha th remaininparameter coul no independentl accoun fo th variance Th goodneso fit o al model wa determine b means o Akaike' (1987 informatio criterio = x2 ~ 2 (df), whic give a superio indicatio o fi imodel wit a smal numbe o parameter (Bollen 1989) an b th chsquar itself Becaus a mode tha ha nonadditiv geneti variancwithou additiv geneti varianc i biologicall implausibl (se Neal &Cardon 1992) th D mode wa neve tested 85OLSON VERNON HARRIS AND JANTabl 2Genetic Analyses of Individual Nonattitude ItemsCategor an itePersonalitHumblAmbitiouFriendlInconsistenPrecisDefensivPersistenFearfuImpulsivSympathetiNeatPlayfuAesthetiObliginDependenInquirinAthleticisAthletiStronAttractivenesPhysica attractivenesGoo lookin t othe peoplAcademiGP i fina yea o hig schooLeve o complete educatioCorrelationM.6.4.3.1.3.3.2.3.4.2.2.3.2.2.3.1.2.2.6.5.3.4.6.8D.2.2.2.0.0.2.2.1.2.0.2.2.1.1.1.0.1.2.2.2.2.2.5ModeAAAADADAAAADAACAADCADAAAAAADACBes fittin modeFiX2(4) = 3.47 nsX2(4) = 0.18 nsX2(4) = 14.72 p .0^(3 = 0.65 ns*2(3 = 1.38 nsX2(4) = 1.42,X2(4) = 2.00 nsX2(4) = 4.06 nsXz(4) = 0.68 nsX*Q) = 0.81 nsX2(4) = 4.25 nsX2(3) = 2.07 nsX2(4) = 1.98 nsX*(4) = 2.57 nsX2(4) = 5.34 nsX2(4) = 1.21 nsX*Q) = 9.30 p .0X2(4) = 9.29 p .0^(3 = 5.43 nsX2(4) = 5.05 nsX2(4) = 3.47 nsX2(4) = 12.87 p .0X2(4) = 10.74 p .0X2(4) = 11.95 p .0^(3 = 0.15 nsX20) = 0.23 nsa2.5.4.0.1.3.3.2.0.0.3.0.3.0.0.0.2.5.5.4.4.4Estimatec2.0.0.0.0.0.0.0.2.1.0.0.0.3e2.4.5.5.7.8.6.6.7.6.5.7.7.6.7.7.5.8.7.7.4.4.5.4.4.1d2.0.2.1.0.

3.2.0.0.3.2.0.0.1Note. Estimate o .0 mea tha th componen wa teste bu di no accoun fo a significan amoun o variance M = monozygoti twins D =dizygoti twins A (a2) = additiv geneti variance E (e2) = nonshare environmenta variance D (d2) = nonadditiv geneti variance C (c2) = shareenvironmenta variance GP = grad poin averageResultGenetic Analyses of Individual ItemsTabl 1 summarize th geneti analyse o th individuaattitud items th intra-twi correlation fo M pair an fo Dpairs th bes fittin model (th ful mode o on o th threreduce models wit it associate ch square an th estimatefo eac o th element i th bes fittin mode (a2, e2, an c2 od2) Becaus onl on o th AC o AD model coul b testefo eac measure eithe c2 o d2 coul no b estimatedth estimate tha wer no teste ar lef blan i th table 2 present th sam statistic fo th individua nonattituditem (personality athleticism attractiveness an academiachievement)A ca b see i Tabl 1 2 o th 3 attitud item yieldesignifican geneti effects becaus o eithe additiv o nonadditiv factor (o both) W summe th additiv an nonadditiv effect t for h2, o heritabilit coefficients Th estimate fo h2 range fro 0 t .57 wit a media o .35 Th fivitem wit th larges geneti component (greate tha o equa t wer attitude towar readin books abortio o demandplayin organize sports rolle coaste rides an th deat penalt murder Th fou item wit th smalles geneti component(n geneti effect wer attitude towar separat role fo mean women playin bingo eas acces t birt control an beinassertiveA ca b see i Tabl 2 1 o th 2 personalit item yieldesignifican additiv and/o nonadditive geneti effects Th estimate fo h2 range fro 0 t .58 wit a media o .29 Th fivitem wit th larges geneti component (greate tha .40 werhumble ambitious exhibitionistic aesthetic an friendly Ththre item wit th smalles geneti component (les tha .20wer neat obliging an inconsistentTabl 2 als show tha bot athleticis item yielde largheritabilit coefficient (greate tha .50) a di bot attractivenes item (greate tha .40) Finally bot measure o academiachievemen als yielde hig heritabilit coefficient (greatetha .40)I i als wort notin tha nonshare environmenta factor(i.e. uniqu experience o eac membe o a twi pair werth mos powerfu contributor t variabilit i bot set oitems Fo items e2 range fro .4 t .75 wit amedia valu o .60 Fo th personalit items e2 range fro.4 t .86 wit a media valu o .71 Th othe nonattituditem (athleticism attractiveness academi achievement alsyielde hig e2, wit th exceptio o leve o complete HERITABILIT O ATTITUDE85education wher e2 fel t .1 (thoug thi wa stil significantl greate tha 0)I shar contras t nonshare environmenta factors shareenvironmenta factor (i.e., experience commo t bot membero a twi pair generall contribute littl t th varianc i thitems Fo th attitud items c2 wa significantl greate tha 0 ionl 1 o th 1 analyse i whic i was tested wit a media o .15 Fo th personalit items c2 wa significantl greatetha 0 i onl 2 o th 1 analyse i whic i was tested wit a valu o 0 Th sam was tru fo th othe nonattituditems fo whic c2 wa significan i onl on o thre analysesFactor Analyses of Individual ItemsA principa component facto analysis wit varima rotationwa conducte o th 3 attitud items Thi analysi yielde nin wit eigenvalue greate tha 1 Tabl 3 present th factoloading o eac ite o eac o th factors Th factor generallseeme interpretabl o th basi o high-loadin items Facto 1reflecte Attitude Towar Athleticism wit hig loading foattitude towar doin athleti activities playin organize sports exercising Facto 2 reflecte Attitude Towar Leadership hig loading fo attitude towar bein th leader bein th o attention publi speaking an bein assertive Facto 3 Attitude Towar th Preservatio o Life wit higloading fo attitude towar abortio o demand voluntar euthanasia eas acces t birt control an organize religio (reversed) Facto 4 reflecte Attitude Towar Sensor Experiences hig loading fo attitude towar lou music smoking rolle rides an bi parties Facto 5 reflecte Attitude TowarIntellectual Pursuits wit hig loading fo attitude towar readin books doin crosswor puzzles playin chess an education towar capitalism als loade highl o thi factor Facto 6 reflecte Attitude Towar Equality wit hig loading foattitude towar makin racia discriminatio illegal separat rolefo me an wome (reversed) open-doo immigratio policiesan gettin alon wel wit others Facto 7 reflecte AttitudeTowar Outwar Appearance wit hig loading fo attitudetowar lookin m bes a al times wearin attention-grabbinclothing an gettin alon wel wit othe people Facto 8 reflecte Attitude Towar th Treatmen o Criminals wit higloading fo attitude towar castratio a a punishmen fo secrime an th deat penalt fo murder Finally Facto 9 wasmor difficul t in

terpret w labele i Attitude Towar Sweetan Games wit hig loading fo attitude towar sweets doin puzzles an playin bingoA principa component facto analysi wa als conducte oth 2 personalit trai items agai wit varima rotation Thi reveale fiv factor wit eigenvalue greate tha 1Tabl 3Factor Analysis of Individual Attitude ItemsAttitudDoin crosswor puzzleDeat penalt fo murdeSweetOpen-doo immigratioDoin athleti activitieVoluntar euthanasiSmokinBein th cente o attentioSeparat role for me an womeEducatio racia discriminatio illega musiGettin alon wel wit othe peoplCapitalis organize sport partiePlayin chesLookin m bes a al timeAbortio o demanPubli speakinPlayin bingWearin clothe tha dra attentioEas acces t birt controExercisin religioBein th leade o groupReadin bookCastratio a punishmen fo se crime assertiv coaste rides1-.0.0-.2.8.0-.3-.0.0-.0.1.2.0.6.2.0.1-.1-.0-.0.7-.0.1-.0.0.22-.1-.0.0.0-.0.7.0-.1.0.0.0.0.0.3.2.0.5-.1.3-.0.1-.0.8.0.5.03-.0.0-.1.0.7.1.0-.2-.0.0.0.0-.0-.0.1.0-.0.7-.1.0.1.6.0-.6.0.1.0.04.0-.0-.1.0.1-.0.5.1-.0-.1-.0.7-.0-.0.2-.0-.0.0.0.2.1-.0-.1.0-.1-.0.5Facto5.5-.1-.0-.0.0.0.0-.1-.0.4.0-.1-.0.4-.1.5-.0.1.3.1-.1.0.0.2.0.6-.0.1-.16-.0-.1.1.4-.0.1-.1-.5.3.7.0-.2-.0.0-.0-.0.1-.0.2.0.0.0.2.0.07-.0.0.0-.1.0-.0-.0.1-.0.3-.0.0.2-.0.3-.1.7.0-.1.0.5.2.1-.0.0.1-.08.1.7-.0-.3-.0.2.2-.0.1.0-.1.1.0.2-.0-.1.1-.0.2-.1-.0-.0.0.0-.1.7.0-.09.4.0.7.2.0.1-.0.1.0-.0.0-.0.0-.2.1.0.0.0.1-.1.4.2-.0-.1.1-.0-.2.0Note. Loading greate tha .4 ar presente i boldface 85OLSON VERNON HARRIS AND JANTabl 4 present th facto loadings Again th factor seemeinterpretable Facto 1 reflecte Aggressiveness wit hig loading fo th trait dominant aggressive exhibitionistic defensive impulsive Facto 2 reflecte Sociability wit hig loading playful friendly aesthetic an caring Facto 3 reflecte wit hig loading fo dependent humble fearfulobliging an defensive Facto 4 reflecte Persistence wit hig fo independent persistent inquiring precise an ambitious Finally Facto 5 reflecte Obsessiveness wit hig loadingfo neat inconsisten (reversed) precise an impulsiv (reversed)Genetic Analyses of Factors and Composite MeasuresTabl 5 summarize th result o th heritabilit analyse o thnin attitud factor tha emerge fro th facto analysis Individuals score fo eac facto wer calculate b weightin al 3item accordin t th facto loading presente i Tabl 3 Tabl summarize th sam statistic fo th heritabilit analyse o thfiv personalit factors usin th weight o th 2 item presente i Tabl 4 Tabl 6 als present th result o heritabilitanalyse o thre supplementar composit measures th su o tw athleticis item (whic correlate significantly r[670 =.60 p .001) th su o th tw attractivenes item (whiccorrelate significantly r[670 = .75 p .001) an th su o standardize value o th GP an complete leve o educatio variable (whic correlate significantly r[670 = .36p .01)A ca b see i Tabl 5 si o th nin attitud factor yieldesignifican additiv and/o nonadditiv geneti components witth estimate fo h2 rangin fro 0 t .66 th media h2 acros thnin factor wa .41 Th thre factor yieldin th larges geneti wer Attitude Towar th Preservatio o Life At Towar Equality an Attitude Towar Athleticism ThTabl 4 Analysis of Individual Personality ItemsPersonalititeHumblAmbitiouFriendlAggressiveInconsistenPrecisPersistenExhibitionistiImpulsivSympathetiNeatPlayfuAesthetiObliginDependen1-.0.3-.0.7-.0.2.5.1.5.0.5-.3-.0.1.0.1.12-.2.2.7-.1.1-.0-.0.1.3-.3.1.5.1.7.5.2.1Facto3.6-.2.1-.2.1.0.4-.0-.0.6.0.3.1-.1.0.5.6.14.1.4.0.7.1.4.1.1.6.0-.0-.0.1.3-.0-.3.55-.2.2.1-.0-.7.5-.1.2.2-.1.1-.4.1.7-.1-.1.1-.0-.1Note. Loading greate tha .4 ar presente i boldfacethre factor yieldin th smalles geneti component wer Attitude Towar Intellectual Pursuits Attitude Towar th Treatmen o Criminals an Attitude Towar Sweet an GamesA ca b see i Tabl 6 thre o th fiv personalit factoryielde significan heritabilit coefficients wit th estimate fo2 rangin fro 0 t .47 Th composit Athleticism Attractiveness an Academi measure als yielde significan genetieffects wit th estimate fo h2 rangin fro .5 t .61Parallelin th analyse o th individua items nonshare environmenta factor constitute th mos powerfu variabl fopredictin varianc i th factor withi ou sample Fo thattitud factors e2 range fro .3 t .63 wit a media valu o.49 Fo th personalit factors e2 range fro .5 t .74 wit amedia valu .62 Analyse o th supplementar composit als yielde significan estimate fo e2, rangin fro.2 t .46Share environmenta factor wer somewha mor importanfo predictin th attitud factor tha the wer fo th individua items wit fou o th seve analyse i whic the wer yieldin a significan estimat fo c2, wit a media valuacros th seve analyse o .26 Tw o th thre analyse o th.personalit factor i whic c2 wa teste yielde significane

stimates wit a media valu o .26 Share experience contrib reliabl t varianc i onl on o th supplementar com measures academi achievementPhenotypic Intercorrelations Among Factors andComposite MeasuresTable 7 an 8 presen th intercorrelation betwee th attitudfactors personalit factors an supplementar composit measure fo M an D twins respectively Becaus o th largnumbe o correlations onl thos correlation tha wer signifi &tp .00 (tw tailed ar presente i boldfac i th tablesRelativel fe correlation wer significan i bot group o (thes correlation ar presente i italic i th tables) W her jus th correlation tha wer significan i bot o twin an tha involve th attitud factorsAttitude Towar Athleticis correlate onl wit self-reporto athleti ability suc tha mor positiv attitude towar athleti wer associate wit greate athleti ability Favorabl atti towar leadershi wer associate positivel wit Aggressiveness Sociability an physica attractiveness bu negativelwit Dependence Attitude Towar th Preservatio o Lif di correlat significantl i bot group o twin wit an personalit o supplementar measures Attitude Towar SensorExperiences fo whic hig score reflecte favorabl attitude intens experiences correlate positivel wit Sociabilitan self-report o athleti abilit but negativel wit Obsessive an academi achievement Favorabl attitude towar intel pursuit wer associate positivel wit Persistence sur suc attitude correlate significantl wit academiachievemen onl withi th M twins Attitude Towar Equalit no correlat significantl i bot group o twin wit anpersonalit o supplementar variables Favorabl attitude towarmaintainin a attractiv outwar appearanc wer associate pos wit Sociabilit an physica attractiveness Attitude To th Treatmen o Criminal (fo whic hig score reflecte fo sever punishment correlate positivel wit Obses HERITABILITY O ATTITUDE85Tabl 5Genetic Analyses of Attitude FactorsAttitude factoAthleticisLeadershiPreservatio o LifSensor ExperienceIntellectua PursuitEqualitOutwar AppearancTreatmen o CriminalSweet an GameCorrelationM.5.6.5.5.5.4.5.3D.3.3.4.4.2.5BesModeAAAACCADACCfittin modeFiX2(4) = 3.84 nsX2(4) = 6.57 nsX2(4 = 3.60 nsX2(3 = 4.64 ns^(4 = 2.59 ns = 2.29 ns�X(4) = 2.51 nsX2(4) = 1.33 nsX2(4) = 8.59 nsa2.5.4.3.0.2.4.0.0Estimatec2.0.0.2.4.0.5.3e2.4.3.3.4.4.6d2.0.2Note. Estimate o .0 mea tha th componen wa teste bu di no accoun fo a significan amoun ovariance M = monozygoti twins D = dizygotic twins A (a2) = additiv geneti variance E (e2) =nonshare environmenta variance D (d2) = nonadditiv geneti variance C (c2) = share environmentavariancesiveness but negativel wit academi achievement Finally Attitude Towar Sweet an Game di no correlat significantl ieithe grou o twin wit an personalit o supplementarmeasuresShared Genetic Variance Between Attitude Factors andPotential MediatorsOn wa o explorin possibl mediator o th heritabilit o avariabl i t calculat th exten t whic geneti variatio ishare wit anothe variable I tw variable shar geneti variance thi mean tha geneti individua difference i on variablar correlate wit geneti individua difference i th othe Suc a patter i consisten wit th ide tha th herita o on variabl mediate o i responsibl fo som o al o heritabilit o th othe (although o course whic variabl i th mediatin i no indicated) Tha is th heritabilit o variabl ma b attributabl t it geneti overla wit th variableT obtai thes statistics w applie Cholesk o trianguladecomposition (se Neal & Cardon 1992 t th data Simila tunivariat heritabilit analyses a Cholesk decompositio assesse whethe th phenotypi correlation withi M twin arhighe tha withi D twins I a Cholesk decomposition however cross-correlation withi twi pair (tha is th correlationbetwee on twin' scor o on o th variable an th othe scor o th othe variable ar examine t estimat thdegre t whic phenotypi correlation ar du t commo ge and/o environmenta factors I th M cross-correlatio i tha th D cross-correlation the a geneti correlatio th tw variable i suggested Tha is geneti individua o on variabl ar correlate wit geneti individua o th othe variableFo eac o th si attitud factor tha yielde significanheritabilit coefficients w calculate th geneti correlatio witeac o th si potentia mediator tha als yielde significan coefficient (thre personalit factor an th threTabl 6Genetic Analyses of Personality Factors and Supplementary Composite MeasuresFacto o measurPersonalit factorAggressivenesSociabilitDependencPersistencObsessivenesSupplementar composit measurAthleticisAttractivenesAcademiCorrelaM.4.2.3.2.6.4.7ionD.2.1.2.2.4.3.4ModeAADAAAACBes fittin modeFi^(4 = 2.43 ns^(3 = 5.71,/i^(4 = 7.26 nsX2(4 = 3.56 nsX2(4 = 4.08 ns/(4 = 13.42 p .0x2(4) = 16.94 p .0x2(4) = 1.23 nsa2.4.1.0.0.6.5.5Estimatec2.2.0.3.0

.0.2e2.5.7.6.3.4.2d2.0Note. Estimate o .0 mea tha th componen wa teste bu di no accoun fo a significan amoun ovariance M = monozygotic twins D = dizygotic twins A (a2) = additiv geneti variance E (e2) =nonshare environmenta variance D (d2) = nonadditiv geneti variance C (c2) = share environmentavariance 85OLSON VERNON HARRIS AND JANTabl 7Intercorrelations Between Attitude Factors, Personality Factors, and Supplementary Composite Measures: Monozygotic TwinsFacto o measurAttitud factorAthleticisLeadershiPreservatio o LifSensor ExperienceIntellectua PursuitEqualitOutwar AppearancTreatmen o CriminalSweet an GameSupplementar composit measur1 Athleticis2 Attractivenes3 AcademiAggressivenes.0.40-.0.2-.0-.2.0.0-.0.2.2-.0Sociabilit.1.22.0.24-.1.1.31-.0.0.24.0Personalit factoDependenc-.1-.17-.1.0.1.0-.0.0-.2-.1.0Persistence.1.1.0-.0.29.1-.0-.1-.0.23.1.23Obsessivenes-.0.0-.1.1.0.1.2.0-.0.0.18Supplementarcomposit measur1.5.1.0.29-.0-.0.0-.1.44.02.1.28.0.1-.0.0.26.0-.0—.13.1.1.0-.16.2.1-.0-.19.0—Note. Correlation presente i boldfac ar significan at .00 (tw tailed) Correlation tha ar significan i bot group o twin appea i italicssupplementar composit variables) Cholesk decompositionals provid a estimat o th exten t whic relation betweevariable ar attributabl t commo environmenta influence(i.e. t th sam nonshare environmenta experience o thtwins) Tabl 9 present th geneti an environmenta correlationfo al pairing o th si attitud factor an th si potentia correlation i boldfac ar significantl greate tha 0Thes dat provid som interestin idea abou wh individuadifference i certai attitude wer heritable whic w elaborat th Discussio section A thi point w simpl summariz th geneti correlations Attitude Towar Athleticis a significan amoun o geneti varianc wit self-reporteathleti abilit an wit Persistence Attitude Towar Leadershishowe geneti overla wit fiv o th si possibl mediatorsphysica attractiveness Sociability Aggressiveness Persistencean academi achievement Attitude Towar th Preservatio oLif produce significan geneti correlation wit physica attrac an Sociability Attitude Towar Sensor Experienceshare geneti varianc wit Sociability Aggressiveness Persis an athleticism Attitude Towar Equalit yielde significan geneti correlation wit Sociabilit an academi achievement Finally Attitude Towar Outwar Appearanc showe overla wit Sociabilit an physica attractivenessTh attitud factor di not i general sho muc overla witth potentia mediator i term o bein attributabl t th sam environmenta experience o th twins Attitude To Athleticis appeare t b influence b th sam environ experience a athleti ability Attitude Towar Leadershi wer associate wit simila environmenta experience aAggressiveness Attitude Towar Sensor Experience wer reTabl 8Intercorrelations Between Attitude Factors, Personality Factors, and Supplementary Composite Measures: Dizygotic TwinsFacto o measurAttitud factorAthleticisLeadershiPreservatio o LifSensor ExperienceIntellectua PursuitEqualitOutwar AppearancTreatmen o CriminalSweet an GameSupplementar composit measur1 Athleticis2 Attractivenes3 AcademiAggressivenes-.0.37-.0.1-.0-.1.0.1.0.1.1-.0Sociabilit.1.17.1.24-.1.2.5.0.0.18-.0Personalit factoDependenc-.0-.0-.1.0.1.1-.1-.1-.0Persistenc.1.1.1-.0.25.0-.1-.1.28.1.29Obsessivenes.0-.1-.36-.0.0.0.28-.0.0.18Supplementarcomposit measur1.43.1.0.17.0-.0.0-.0.31.02.0.25.1-.0.0.21-.0.0.03.1.1.0-.24.1.1-.1-.20-.0—Note. Correlation presente i boldfac ar significan a p .00 (tw tailed) Correlation tha ar significan i bot group o twin appea i italics HERITABILITY O ATTITUDE85Tabl 9Genetic and Environmental Correlations Between Attitudesand Potential MediatorsAttitude facto anmediatoAthleticisAggressivenesSociabilitPersistencAttractivenes achLeadershiAggressivenesSociabilitAthleticisAttractivenesAcademi ach.Preservatio o LifAggressivenesSociabilitPersistenceAthleticisAttractivenes achSensor ExperienceAggressivenesSociabilitAthleticisAttractivenesAcademi achEqualitAggressivenesSociabilitPersistenceAthleticisAttractivenesAcademi ach.Outwar AppearancAggressivenesSociabilitAthleticisAttractivenesAcademi ach.h2.5.4.3.6.5.5.4.4.4.3.6.5.5.4.4.3.6.5.5.3.4.4.3.6.5.5.6.5.5.4.4.4.3.6.5.5Geneticorrelatio-.0.0.2.6.1.4.4.2.1.5-.1.2.0.2-.0.3-.2.2.1-.0-.1.4.1.0.1.0.3-.1-.0.2-.0Environmentacorrelatio.0.1.1.3.0.0.3.0.1-.0.1-.1.0.0-.0.0.0.1.2.0-.1-.1-.0.0-.0-.0.0.3.0.0.2.0Note. Correlation presente i boldfac ar significan a p .00 (twtailed) h2 = heritabilit coefficients ach = achievementlate t th sam environmenta experience a athleticism Attitude Towar Equalit showe a negativ environmenta correla wit academi achievement whic mean tha environmenta tha wer associate wit positiv attitude towar als tende t b negati

vel associate wit academiachievement Finally Attitude Towar Outwar Appearanc relate t th sam environmenta experience a wer Sociabilit an physica attractivenessTesting Phenotypic Mediation of Attitude FactorsAnothe mean o examinin mediatio relie o phenotypicorrelations testin directio o causalit model tha g fro thphenotyp o on variabl (e.g. a personalit variable t thphenotyp o anothe variabl (e.g. a attitude) Technique hav develope tha allo th testin o simpl model o causatio usin cross-sectiona famil dat (se Care & DiLalla 1994 e al. 1993) Lik th Cholesk analyse previousl described thes technique us cross-correlation withi M an Dtwi pairs Fo example i th personalit trai o sociabilit positiv attitude towar parties the th cross-correlatio on twin' Sociabilit an th othe twin' attitud towarpartie shoul b equa t th produc o tw othe correlations thintraindividua correlatio betwee Sociabilit an attitude towar partie an th correlatio betwee on twin' Sociabilitan th othe twin' Sociability Wit sufficien sampl sizes thobtaine cross-correlatio ca b teste agains th predicte value a significan differenc suggest tha th mode i invalid.Becaus w wer intereste specificall i th mediatio ogeneti effects w conducte thes phenotypi analyse onl fo pairing tha produce a significan geneti correlatio i Tabl 9 (i.e. w use thes analyse a anothe wa oexplorin whether a attitud migh b heritabl becaus i i b anothe variabl tha i heritable) O th 1 sucpairings 8 di no produc a significan phenotypi correlatiowithi bot set o twin i Table 7 an 8 clearly i mad nsens t tes phenotypi mediatio whe ther wa n phenotypicorrelatio betwee th tw variablesThus ther wer nin attitude-mediato pair tha produce bota significan geneti correlatio an a significan phenotypi correlation Fo thes pairs w teste acros al twin (bot M anD t increas power tw contrasting simpl model o causa th mediato cause th attitud versu th attitud cause th Unfortunately si o th nin test yielde inconclusiv eithe becaus th tw model yielde ver simila predic withi ou sampl o becaus th correlatio betwee thpotentia mediato an th attitud wa smal (o both) Fo thessi pairs neithe causa mode coul b rejecte o th basi o oudataO th othe hand thre pair o variable yielde result thasupporte on mode ove th other First ou dat supporte thconclusio tha athleti abilitie caus Attitude Towar AthleticsTh predicte cross-correlatio fo thi model wa r = .29 whic no diffe significantl fro th obtaine cross-correlation r =.3 (cross-correlation ca b compute i tw ways fro Twi 1t Twi 2 o fro Twi 2 t Twi 1 s w average th twvalues) I contrast th predicte cross-correlatio fo th modeassumin tha Attitude Towar Athletic caus athleti abilitiewa r = .23 whic differe significantl (p .05 fro th valu o .34Unexpectedly ou dat weakl supporte th conclusio thaAttitude Towar Sensor Experience caus th personalit variable o Aggressiveness an Sociability Wit regar t Aggres th Aggressiveness cause attitude mode predicte a o r = .07 whic differe significantl fro thobtaine cross-correlatio o .17 th attitude caus Aggressive mode predicte a cross-correlatio o .11 whic di no significantl fro .1 (thoug i was no muc large tha precedin correlatio o .07) Wit regar t Sociability thSociabilit cause attitude mode predicte a valu o r = .07whic differe significantl fro th obtaine cross-correlatio o.16 th attitude caus Sociabilit model predicte a crosscorrelatio o .13 whic di no diffe fro .16 Thus fo thes 85OLSON VERNON HARRIS AND JANtw pairings th phenotypi dat wer mor consisten wit thhypothesi tha th attitude cause th hypothesized mediato tha th revers causa hypothesisPredicting the Heritability Coefficients of Attitude ItemsA a fina wa o explorin possibl mediator o th heritabilitie o attitudes w examine th correlation betwee (a thheritabilit coefficient o th attitude an (b th absolut valueo th correlation betwee th attitude an othe variable (potential mediators) Fo example acros th 3 attitud items wcalculate th correlatio betwee th heritabilit coefficient o 3 item (h2) an th 3 absolut value o th correlationbetwee th item an th personalit facto o Aggressiveness Apositiv correlatio woul indicat tha a attitud item correlate strongl wit Aggressiveness (eithe positivel o negatively absolut value wer computed) the (attitud items alstende t b mor heritable I othe words a positiv correlatiowoul b consisten wit th vie tha i th strengt association wit Aggressiveness accounte fo som o th i th heritabilitie o th attitud items Aggressiveneswoul b implicate a a possibl mediato o th heritabilit oth attitude measure i ou study Thi analysi assume tha al th measure attitude hav th sam underlyin determinan o o thei heritability i i quit possibl (o evelikely tha differen attitude ar

heritabl fo differen reason(i.e., hav differen mediator o thei heritability) whic woulno b reveale b thi particula statistica approach.W compute thes correlation fo th mediator tha werthemselve heritabl (thre personalit factor an th thre supplementar composit measures) Onl on variabl yielde asignifican result Specifically th amoun o geneti varianc i attitud (h2) wa positivel correlate wit th correlatiobetwee th ite an th academi achievemen composit measure r(28 = .33 p .04 Thus attitude tha correlate witacademi achievemen tende t b mor heritable Give tha achievemen wa itsel highl heritabl an correlate ath geneti leve wit som attitude i th earlier Choleskdecompositio analyses thi correlatio wit th heritabilit o i provocativeRelation Between Attitude Heritability and AttitudeStrengthImmediatel afte reportin thei attitud towar a target participant indicate ho importan thi attitud wa t the an strongl the hel th attitude Thes question assesse strength whic Tesse (1993 ha argue migh b relatet attitud heritability T examin thi hypothesis w calculate eac attitud ite th mea ratin o importanc acros alparticipant an th mea ratin o strengt acros al participantseac o thes tw set o mean wa the correlated acros th 3items wit th heritabilit coefficient o th items Mea ratingo importanc di not correlat wit h2 acros th 3 itemsr(28 = .05 ns. Althoug th relatio wa no statisticall significant mea rating o strengt wer positivel correlate wit h2acros th 3 items r(28 = .17 p .19 a predicte b Tesse(1993)W als looke a th relatio betwee heritabilit an strengtwithi th attitud factors Thus w compute a weighte impor scor an a weighte strengt scor fo eac attitud factob multiplyin th mea importanc (o strength ratin fo eacite b th facto loading i Tabl 3 I s doing w generate singl importanc (an strength scor fo eac o th nin W the correlate thes nin score wit th heritabilitcoefficient o th attitud factors More-heritabl factor werstrongl associate wit bot highe weighte importanc scores = .52 p .08 an highe weighte strengt scores r(7 =.51 p .08 althoug th correlation wer onl marginallsignifican becaus o th fe degree o freedom Thus a predicte b Tesse (1993) heritabl attitud factor tende als t bpsychologicall strongerDiscussioThi stud provide stron evidenc tha difference betweerespondent i man o thei expresse attitude wer partl de b geneti factors Twenty-si o th 3 individua atti item yielde significan geneti effects wit a media h2 o.35 Si o th nin attitud factor yielde significan genetieffects wit a media h2 o .41 Thes geneti effect emergeacros a wid variet o attitud topic an domains fro attitude divers a suppor fo th deat penalt t enjoymen o rollecoaste ridesOu result ar consisten wit pas studie o th heritabilit osocia attitude (e.g. Arve e al. 1989 Eave e al. 1989 Kellee al. 1992 Walle e al. 1990) whic hav als foun a geneticomponen i man evaluations Ou stud include man ne topic neve previousl teste i twi research suc aattitude towar physica activities intellectua tasks an leadershi behaviors Difference betwee respondent i som o thesne attitude prove t b highl heritabl (e.g. Attitude TowarAthleticis an Attitude Towar Leadership)Wha abou th specifi attitude tha hav yielde hig o loheritabilitie i pas studies A state i th introduction on oou goal wa t tes th replicabilit o th heritabilit coefficientfo particula attitude i pas research I general attitude thahav yielde hig heritabilitie i pas studie als di s i thi Fo example lik i Eave e al (1989) a substantiageneti componen wa foun i participants attitude towar thdeat penalt fo murde an voluntar euthanasia Lik Walle eal (1990) w foun tha attitude towar organize religio werals heritable On puzzlin exceptio t thi patter wa thattitud facto o th treatmen o criminal ( domai tha ha hig heritabilitie i th past) whic wa no heritabl th fac tha th tw highes loadin items attitude towarth deat penalt an castratio fo se crimes yielde significan coefficients Perhap som item wit lowe wer no heritabl (e.g. attitude towar bingo o tha wer but conceptuall distinc fro Attitude Towar th o Criminal (e.g. attitude towar smoking maske heritabilit o th highes loadin items.' Thes dat rais a issu concernin ou heritabilit analyse o thattitud factors Give tha th facto analyse wer conducte o thphenotypi covariances item tha loade heavil o th sam facto di HERITABILIT O ATTITUDE85W wer les successfu a replicatin previou finding o loheritabilities Attitude towar separat role fo me an womewer no heritable replicatin previou finding fo simila itemdealin wit gende role (Eave e al. 1989) Severa othe tha hav yielde lo heritabilitie i pas research wer mor heritabl i ou sample Fo example heritabilit coefficient o .54 .46 .39 an .3 wer obtaine

foattitude towar abortio o demand open-doo immigration cap an smoking respectively whic hav yielde lo o estimate i pas studie (Eave e al. 1989 Perry1973) Whe al o th relevan studies includin th presen onear examined ther ar fe attitud topic tha hav consistentl nul heritabilit estimate (although o course man possibl attitud topic hav ye t b examine i behaviora geneticresearch)W shoul th reade o som o th assumption underlyin th twi methodolog adopte i thi study First w tha ou zygosit determinatio wa accurate—tha is thquestionnair designe b Nichol an Bilbr (1966 approacheth accurac rat o 93 tha ha bee reporte i pas studiesSecond w assume tha th degre o environmenta similaritwa equa fo identica an fraterna twins Thi assumptio hagenerall bee supporte b empirica test (e.g. se Plomi e al.1997) althoug i remain contentiou fo critic o behavioragenetics Violatio o thi assumptio (wit identica twin bein t mor simila environments woul ten t increasheritabilit estimates Third w assume tha relativel littl assortativ matin occurre i ou sample Assortativ matin i thtendenc fo geneticall simila individual t mate whic wil geneti similarit withi familie an decrease genetisimilarit betwee families Becaus suc increase intrafamilia similarit wil occu fo D twin but no M twin (wh geneticall identica irrespectiv o assortativ mating) stron matin woul ten t decreas heritabilit estimatesFinally give tha w wan t generalize th finding fro oustud t th genera population w assume tha th sampl otwin wa reasonabl representativ o th large population T exten tha th heterogeneit o th gene and/o environment ou sampl differe substantiall fro thos i th large population ou estimate o geneti an environmenta influenc wilno generalizeMediators of Attitude HeritabilityI th conclusio i accepte tha man attitude ar partlheritable the th questio arises wh o ho ar difference individuals attitude geneticall determined Give tha gene-to-attitud connection ar extremel unlikely wha th mechanism tha migh accoun fo th geneti componen attitudes Thi issu ha receive ver littl attentio fro scientists W explore i i severa way i th presenresearchMos directly w calculate th exten t whic geneti variatio wa share betwee heritabl attitud factor an heritablno necessaril sho simila pattern o heritabilitie themselves Aalternativ strateg migh hav bee t conduc a facto analysi o th covariance tha derive simpl fro share geneti variancepotentia mediator (namely personalit factor an supplementar composit variables) Geneti correlation indicat th extent whic geneti individua difference i on variabl ar relatet geneti individua difference i anothe variable Causa directio i no reveale b geneti correlations o course a poin tha illustrat severa time b presentin competin possibl interpretation o ou results Nevertheless a describe i th followin paragraphs ou geneti analyse identifie severa potentia tha overlappe considerabl wit attitude (a wel a smalle bu significan geneti correlation tha w d nodiscuss) Al o th mediator yielde significan geneti correlation wit a leas tw o th attitud factors th personalit trai Sociabilit wa particularl impressive correlatin a th ge- leve wit fiv o th si attitud factors W followe u th correlatio analyse wit phenotypi mediatio analyse tw simpl models th mediato cause th attitud versuth attitud cause th mediator Thes analyse onl mad sens th variable wer correlate a th phenotypi leve and then wer usuall inconclusiv becaus o relativel wea correlations A fe mediator-attitud pairings however yielde interestin dat tha ar als discusse belowAttitude Towar Athleticism whic wer heritabl (h2 = .54)showe a hig geneti correlatio (.63 wit self-reporte athleti whic wa itsel highl heritabl (h2 = .61) On interpretatio o thi overla i tha inherite physica abilitie relevan t" athletic influence attitude towar athletic (playin organizesports exercising etc.) Fo example individual wit naturallgoo coordinatio an strengt migh hav bee mor successfu sport tha les athleticall incline individuals wit th resultha th forme individual develope mor favorabl attitudetowar sport tha di th latter Thi causa mode wa explicitlsupporte b th phenotypi analyses fo whic th mode goinfro athleti abilitie t athleti attitude wa consisten wit thdata wherea th mode goin fro athleti attitude t athletiabilitie coul b rejected Thus ther wer convergin piece o i ou stud supportin th conclusio tha attitude athletic yielde a significan heritabilit coefficient a i part becaus the wer cause b geneticall influence an athleti abilitiesAttitude Towar Leadershi (h2 = .41 showe relativel larggeneti correlation wit self-reporte physica attractivenes(.55) Sociabilit (.43) an Aggressiveness (.41) al o whic als highl heritabl (h2 =

.54 .47 an .43 respectively)Thus i i possibl tha Attitude Towar Leadershi wer heritable i part becaus othe inherite characteristic le t particulaexperience relevan t leadership Fo example perhap physi attractiv person wer treate deferentiall an developemor confidenc i thei leadershi abilities Similarly perhap wh wer naturall outgoin (sociable gravitate tposition o leadershi an develope positiv attitudes Naturaaggressivenes o assertivenes migh als hav le t greate i attempt t lead whic induce favorabl attitudetowar bein a leader Thes variable constitut a goo exampl th ambiguit o geneti correlations however i tha the ca b interprete i othe way (phenotypi mediatio analyse involvin Attitude Towar Leader wer inconclusive) Foexample Attitude Towar Leadershi ma hav bee heritablfo reason othe tha thos w hav proposed an th heritabil o th hypothesized mediator ma themselve hav bee du 85OLSON VERNON HARRIS AN JANt th heritabilit o Attitude Towar Leadership Fo exampleperhap individual wh wer favorabl towar leadershi madmor effor t mak themselve physicall attractive o becammor sociabl an aggressiv becaus the wante t b leadersAttitude Towar Sensor experience (h2 = .36) wit highescore reflectin mor positiv attitude towar intens experience lik lou music bi parties an rolle coaste rides, share substantia amoun o geneti varianc wit Sociabilit (r = .41)Perhap th trai o sociabilit encompasse a genera tendenc tapproac source o stimulatio (includin othe people) whic i positiv attitude towar variou kind o sensor Thi relatio migh hav reflecte th opposit causa however Fo example positiv attitude towar intenssensor experience ma hav induce th trai o sociabilit bmotivatin peopl t approac socia gatherings Indeed th phe causa analyse supporte thi latte interpretation Th wer consisten wit th causa model goin fro sensorattitude t Sociability wherea th revers causa model (whic th on w woul hav predicted wa disconfirme b thAttitude Towar Sensor experience als yielde a moderatel hig geneti correlatio wit Aggressiveness (.33) Perhap aggressivenes induce relativel intens sensor expe whic produce mor positiv attitude towar suc ex Onc again though th phenotypi analyse wer morconsisten wit a opposin view Th causa mode goin fro attitude t Aggressiveness wa supporte b th datawherea th mode goin fro Aggressiveness t attitude wadisconfirmed Thus fo example positiv attitude t intenssensor experience (heritabl fo whateve reason ma havfacilitate th developmen o a aggressiv personalit b moti intens experience lik interpersona confrontation o competitive activitiesAttitude Towar Equalit (h2 = .55 yielde a stron geneticorrelatio (.44 wit Sociability Thi overla i intriguing be th attitud item include i th equalit facto reflected a i part a desir fo positiv interpersona relationship (e.g. racia discriminatio illegal gettin alon wel wit othepeople) Thus natura inclination t b sociabl migh hav a desir fo smoot interactions whic affecte attitude issue involvin interpersona relationships O course i i possibl tha positiv attitude towar gettin alon wit othe coul hav facilitate th developmen o sociabilit (phenotypi mediationa analyse wer inconclusive)Th fina heritabl attitud factor Attitude Towar OutwarAppearanc (e.g. lookin m bes a al times wearin attention clothes h2 = .45) yielde a moderatel hig geneticorrelatio wit Sociabilit (.36) Natura inclination towar so coul plausibl hav induce mor interes i lookin (t impres o attrac others) Alternatively thi relatiomigh hav reflecte tha positiv attitude towar lookin gooincrease Sociabilit (phenotypi analyse wer inconclusive)W conducte another ver differen analysi tha explore thexten t whic th potentia mediator coul accoun fo th o man attitudes Specifically w calculate th betwee th heritabilit coefficient o th 3 attitud an th items correlation wit th mediators Rathe tha geneti overla betwee a specifi attitud an possibl thi analysi investigate th abilit o th mediator taccoun fo difference betwee th attitude i heritability Onlon mediato yielde a significan result—th academi achievemen measure Th correlation betwee academi achievemen eac o th attitud item wer significantl relate (r = .33t th heritabilit coefficient o th attitudes Thus attitud item correlate highl wit academi achievemen als tende t bhighl heritable (I a simila vein ther i als a positiv correlatio betwee th g loading o I subtest an th subtests se Jensen 1998. Ou resul suggest tha academiachievemen ma hav accounted a leas i part fo th heritabilit o individua difference i man attitudes Tha is academiachievement whic wa itsel heritable ma hav contribute tth heritabl componen o individua difference i man attitudes Thi hypothesi i consisten wit th conclusion o Scaran We

inber (1981) wh argue tha th geneti transmissio overba abilit (on probabl aspec o academi achievement accounte fo th geneti componen o individua differ i authoritaria attitudes W shoul reemphasize however thi analysi rest o th assumptio tha th sam variabl th heritabilit o man attitudes whic arguabl i a assumption Also academi achievemen migh havreflecte respondents motivatio a muc a (i no mor thanthei intellect i whic cas motivationa variable migh havbee th operativ variabl influencin attitude acros mandomainsEnvironmental Effects on AttitudesW hav focuse t thi poin o th relevanc o ou dat founderstandin th geneti transmissio o individua difference iattitudes bu i i importan t emphasiz tha ou finding als confirme th importanc o environmenta variable i formation I al o ou analyses th larges percentage o i attitude wer attributabl t nonshare environmentafactors Tha is th mos significan predicto o variabilit iattitud score (whethe a th ite o facto level wa variatio i tha wer uniqu t individua member o twi pairs course thes finding shoul no b surprisin t socia psychologists wh subscrib strongl t a belie i environmenta Nevertheless i th contex o a articl o gene anattitudes i i wort keepin i min th fac tha environmentafactor outweighe geneti one a predictor o th attitude iou sampl (althoug w shoul als not tha estimate fo non environmenta factor wer calculate a residual tha varianc du t issue suc a ite unreliability whic b considere erro rathe tha environmenta influence)Th observatio tha nonshare environmenta factor ar importan i no ne t ou research Indeed behaviora geneticist increasingl bee notin bot th predictiv strengt o experience an th surprisin weaknes o share envi factor i th domain o personality attitudes anaptitude (se Dun & Plomin 1990 Eave e al 1989) Foexample Plomi an Rend (1991 stated "On o th mos discoverie i huma behaviora genetic involve nur rathe tha nature Environmenta factor importan t devel ar experience differentl b childre i th sam family(p 179) A importan directio fo futur researc i t explormor thoroughl th specifi experience o sibling tha lea t i attitudes Suc researc ma requir bette measure HERITABILITY O ATTITUDE85o th environmen tha ar currentl availabl (Plomi & Rende1991) althoug i a recen stud usin multipl environmentameasures th author ha som succes identifyin nonshareenvironmenta mediator o personalit difference (Vernon Jang & McCarthy 1997)Th Cholesk analyse o ou dat identifie a fe attitudemediato pair tha share environmenta origins Tha is i a fecases a attitud facto an a potentia mediato variabl yieldea significan environmenta correlation indicatin tha simila experience influence bot variables I al o th case a significant positiv environmenta correlation th tw vari als correlate positivel a th phenotypi level Specifi simila experience wer implicate i th developmen obot variable o (a athleti abilit an Attitude Towar Athleticism (b Aggressivenes an Attitude Towar Leadership (cathleti abilitie an Attitude Towar Sensor Experiences (dSociabilit an Attitude Towar Outwar Appearance an (ephysica attractivenes an Attitude Towar Outwar Appearance Fo example exposur t a goo coac migh bot improvathleti abilitie an foste positiv attitude towar organizesport (th firs environmenta correlatio liste above) O earlexperience o pee ridicul migh bot produc a shy nonaggressiv styl an precipitat negativ attitude towar bein a leade secon environmenta correlation) Thes environmenta find furthe underscor th importanc o studyin th specifi o sibling tha lea t particula psychologicacharacteristicsAttitude Heritability and Attitude StrengthTesse (1993 hypothesized tha heritabl attitude ar psychologicall stronge tha ar attitude tha ar no heritabl (se alsCreli & Tesser 1996 Tesse & Crelia 1994) H showe thaattitude tha ha yielde hig heritabilitie i pas researc wer mor quickly wer les influence b conformit pressure affecte interpersona likin more an yielde strongeagreement-reinforcemen effect tha di attitude tha ha lo heritabilities Tesse speculate tha highl heritabl ma hav a biologica basi tha make chang difficul ouncomfortable thus individual buil u psychologica defensearoun suc attitude an see ou o construc environment thasuppor an protec themN researcher hav previousl teste thi hypothesi b obtainin direct self-repor measure o attitud strengt fro participants I th presen study w obtaine suc measure an exam-ine thei relatio t attitud heritability Thi approac yielde tha supporte Tesser' (1993 reasoning Specifically th 3 attitud items ther wa a wea correlatio o .1 wa statisticall nonsignificant betwee th heritabilit oth ite an th mea ratin acros al participant o ho strongl hel th attitude Item tha wer mor heritabl als tende

yiel somewha highe mea strengt rating acros participantsMor impressively whe mea importanc an mea strengtscore wer calculate fo eac attitud facto (b weightin thimportanc o strengt mean fo th 3 item b th factoloading i Tabl 3) thes score wer strongl relate t th o th factor (correlation o .5 an .51) althougth correlation wer onl marginall significan becaus o th degree o freedom Twenty-fiv percen o th variatio i thstrengt o importanc score o th attitud factor coul baccounte fo b variatio i th heritabilitie o th factorsThes dat provid suppor fo Tesser' (1993 hypothesi thahighl heritabl attitude ar stronge tha ar les heritabl attitudes Moreover ou suppor come fro a entirel differenresearc strateg tha thos use b Tesser Ther ar a variet oway tha attitud strengt ca b operationalize (se Bassili1996) ou findings togethe wit thos o Tesse (1993 Creli &Tesser 1996 Tesse & Crelia 1994) provid convergin evidenc ver differen operationalization tha highl heritabl atti ar psychologicall stronge tha les heritabl attitudesConclusionsAttitude ar learned Bu attitude als depen o biologicafactors I thi research w obtaine evidenc tha variatio acrosindividual o a wid variet o attitude i partl attributabl tgeneti factors Th media h2 acros th 3 individua attituditem wa .35 meanin that typically 35 o th attitudinavarianc i ou sampl wa attributabl t geneti factors Non environmenta factor wer eve mor importan i predictin variatio withi ou sample producin a media e2 o .6acros th 3 individua attitud items Nevertheless full 2 o 3 individua attitude yielde a estimat o geneti influenc wa significantl greate tha zeroSevera personalit trait an supplementar variable werimplicate a possibl mediator o attitud heritability Presum thes characteristics—themselve highl heritable—predispose individual t for particula kind o attitudes therebcontributin t th geneti determinatio o individua difference thos attitudes Sociability i particular wa correlate a ageneti leve wit severa attitudes Athleti abilit an physica als yielde substantia geneti correlation wit attitudes Finally academi achievemen appeare t b o predictin th heritabilit o th se o attitude measure i ou sample W thin tha th primar valu o oumediationa analyse i t rais interestin question abou thsource o attitud heritabilit tha ca b examine i futurresearchW hop tha attitud researcher wil begi t direc morattentio t th rol o biologica processes includin genetifactors i th formatio an chang o attitudes Twi methodol constitut jus on o a variet o approache tha offe th t examin th simultaneou effect o natur an nurture I th lon run w stan t gai th mos understandin froperspective tha integrat biolog an experienc i accountinfo individua differencesReferenceAdorno T W. Frenkel-Brunswik E. Levinson P J. & Sanford R N(1950) The authoritarian personality. Ne York Harpe & RowAkaike H (1987) Facto analysi an AIC P.sychometrika, 52, 317-332Arvey R D. Bouchard T }. Segal N L. & Abraham L M (1989) Josatisfaction Environmenta an geneti components Journal of AppliedPsychology, 74, 187-192Bassili J N (1996) Meta-judgmenta versus operativ indexe o psychologica attributes Th cas o measure o attitud strength Journal of and Social Psychology, 71, 637-653 86OLSON VERNON HARRIS AN JANBollen K A (1989) Structural equations with latent variables. NeYork WileyBostrom R N. & White N D (1979) Doe drinkin weake resistanceJournal of Communication, 29, 73-80Bouchard T J Jr. & McGue M (1990) Geneti an rearin environmenta influence o adul personality A analysi o adopte twinreare apart Journal of Personality, 58, 263-292Cacioppo J T. & Petty R E (1987) Stalkin rudimentar processe osocia influence A psychophysiologica approach. I M P Zanna J MOlson & C P Herma (Eds.) Social influence: The Ontario symposium(Vol 5 pp 41-74) Hillsdale NJ ErlbaumCarey G. & DiLalla D L (1994) Personalit an psychopathologyGeneti perspectives Journal of Abnormal Psychology, 103, 32-43Crelia R A. & Tesser A (1996) Attitud heritabilit an attitudreinforcement A replication Personality and Individual Differences,21, 803-808DeFries J C Plomin R. & LaBuda M C (1987) Geneti stabilit ocognitiv developmen fro childhoo t adulthood DevelopmentalPsychology, 23, 4-12Dunn J. & Plomin R (1990) Separate lives: Why siblings are sodifferent. Ne York Basic BooksEagly A H. & Chaiken S (1993) The psychology of attitudes. SaDiego CA Harcour Brac JovanovichEaves L J. Eysenck H J. & Martin N G (1989) Genes, culture, andpersonality: An empirical approach. London Academi PressEysenck H J (1951) Primar socia attitude a relate t socia clas anpolitica party British Journal of Sociology, 2, 198-209Heath A C Kessler R C Neale M C Hewitt J K. Eaves L J. &Kendler K S (1993) Testin hypothese

abou directio o causatiousin cross-sectiona famil data Behavior Genetics, 23, 29-50Jackson D N (1984) Personality Research Form manual (3r ed.) PorHuron MI Researc Psychologist PressJensen A R (1998) The g factor: The science of mental ability. WestportCT PraegerJockin V. McGue M & Lykken D T (1996) Personalit an divorceA geneti analysis Journal of Personality and Social Psychology, 71,288-299Joreskog K G. & Sorbom D (1989a) LISREL: A guide to program andapplications (2n ed.) Mooreseville IN Scientifi SoftwareJoreskog K G. & Sorbom D (1989b) PRELIS: A preprocessor forLISREL (2n ed.) Mooreseville IN Scientifi SoftwareKasriel J. & Eaves L (1976) Th zygosit o twins Furthe evidenc oth agreemen betwee diagnosi b bloo group an writte questionnaires Journal of Biosocial Science, 8, 263-266Keller L M. Bouchard T J Jr. Arvey R D. Segal N L. & DawisR V (1992) Wor values Geneti an environmenta influencesJournal of Applied Psychology, 77, 79-88Loehlin J C (1989) Environmenta an geneti contribution t behaviora development American Psychologist, 44, 1285-1292Loehlin J C & Nichols R C (1976) Heredity, environment, andpersonality: A study of 850 sets of twins. Austin Universit o TexaPressMartin N G. Eaves L J. Heath A C Jardine R. Feingold L R &Eysenck H J (1986) Transmissio o socia attitudes Proceedings ofthe National Academy of Sciences, USA, 83, 4364-4368Neale M C & Cardon L R (1992) Methodology for genetics studies oftwins and families. Norwell MA Kluwe AcademicNichols R C & Bilbro W C Jr (1966) Th diagnosi o twi zygosityAda Genetica Medicae et Gemellologiae, 16, 265-275Olson J M. & Zanna M P (1993) Attitude an attitud change AnnualReview of Psychology, 44, 117-154Oskamp S (1991) Attitudes and opinions (2n ed.) Englewoo CliffsNJ Prentic HallPartanen J. Bruun K. & Markkanen T (1966) Inheritance of drinkingbehavior. Helsinki Finland Th Finnis Foundatio fo AlcohoResearchPerry A (1973) Th effec o heredit o attitude towar alcoholcigarettes an coffee Journal of Applied Psychology, 58, 275-277Plomin R. Corley R. DeFries J C & Fulker D W (1990) Individuadifference i televisio viewin i earl childhood Natur a wel anurture Psychological Science, 1, 371-377Plomin R. DeFries J C & Loehlin J C (1977) Genotype-environmeninteractio an correlatio i th analysi o huma behavior Psycho-logical Bulletin, 84, 309-322Plomin R. & Rende R (1991) Huma behaviora genetics AnnualReview of Psychology, 42, 161-190Scarr S. & McCartney K (1983) Ho peopl mak thei ow environments A theor o genotype-to-environmen effects Child Develop- 54, 424-435Scarr S. & Weinberg R A (1981) Th transmissio o authoritarianisi families Geneti resemblanc i social-politica attitudes I S Scar(Ed.) Race, social class, and individual differences in I. Q. (pp 399427) Hillsdale, NJ ErlbaumSnyderman M. & Rothman S (1987) Surve o exper opinio ointelligenc an aptitud testing American Psychologist, 42, 137-144Tellegen A. Lykken D T. Bouchard T J Jr. Wilcox K J. Segal N L.& Rich S (1988) Personalit similarit i twin reare apar antogether Journal of Personality and Social Psychology, 54, 1031-1039Tesser A (1993) Th importanc o heritabilit i psychologica researchTh cas o attitudes Psychological Review, 100, 129-142Tesser A. & Crelia R A (1994) Attitud heritabilit an attitudreinforcement A tes o th nich buildin hypothesis Personality and Differences, 16, 571-577Vernon P A. Jang K L. Harris J A. & McCarthy J M (1997)Environmenta predictor o personalit differences A twi an siblinstudy Journal of Personality and Social Psychology, 72, 177-183Waller N G. Kojetin B A. Bouchard T J Jr. Lykken D T. &Tellegen A (1990) Geneti an environmenta influence o religiouinterests attitudes an values A stud o twin reare apar an together Psychological Science, 1, 138-142Wilson G D. & Patterson J R (1968) A ne conservatis scale BritishJournal of Social and Clinical Psychology, 7, 264-269Zanna M P. & Cooper J (1974) Dissonanc an th pill A attributioapproac t studyin th arousa propertie o dissonance Journal ofPersonality and Social Psychology, 29, 703-709Zuckerman M (1995) Goo an ba humors Biochemica base opersonalit an it disorders Psychological Science, 6, 325-332Receive Apri 2 199Revisio receive Marc 1 199Accepte Decembe 15 199 • ATTITUDE AN SOCIA COGNITIOTh Heritabilit o Attitudes A Stud o TwinJame M Olson Phili A Vernonan Juli Aitke HarriUniversit o Wester OntariKerr L JanUniversit o Britis ColumbiTh geneti basi o individua difference i attitude wa examine i a surve o 19 pair omonozygoti twin an 14 pair o same-se dizygotic twins A principa component analysi o th 3attitud item i th surve identifie 9 attitud factors o whic 6 yielde significan heritabilitcoefficients Nonshare environmenta factor accounte fo th m