Mate Choice and Friendship in Twins.

Research Article

Mate Choice and Friendship in Twins Evidence for Genetic Similarity J. Philippe Rushton and Trudy Ann Bons

The University of Western Ontario, London, Ontario, Canada

ABSTRACT—This study examined the genetic and environ-

mental contribution to people’s preference for spouses and

friends to be similar to themselves. In their responses to 130

personality, attitude, and demographic questions, 174

pairs of monozygotic (MZ) twins resembled each other (r 5

.53) more than did 148 pairs of dizygotic (DZ) twins (r 5

.32), 322 pairs of spouses (r 5 .32), and 563 pairs of best

friends (r 5 .20). It was not previously recognized that

spouses and friends are as similar as DZ twins. MZ twins

also chose spouses and best friends more similar to their co-

twins’friendsandspousesthandidDZtwins(meanrs5.22

vs. .14). The twins’ preference for spouses and friends

similar to themselves was about 34% due to the twins’

genes, 12% due to the twins’common environment,and 54%

due to the twins’ unique (nonshared) environment. Similar-

ity to partners was more pronounced on the more heritable

items than the less heritable items. It is concluded that peo-

ple are genetically inclined to choose as social partners

those who resemble themselves at a genetic level.

As the English language makes obvious, ‘‘likeness goes with

liking.’’Similarity,whetheractualorperceived,isoneofthemost

important variables in all of social psychology, influencing in-

terpersonal processes from attraction to grouping and from per-

suasion toprejudice.People’spreference for theirsocial partners

to be similar to themselves is usually attributed to cultural and

cognitive factors, such as being raised in a particular way or

confirming one’sjudgmentsabouttheworld.Evolutionary factors

may also be at work through genes and sexual imprinting (Be-

reczkei, Gyuris, & Weisfeld, 2004; Rushton, 1989a).

In a study of 1,000 sexually interacting couples of northern

European appearance (judged by photographs), couples who

produced a child together were 52% similar with respect to 10

blood groups, whereas those who did not were only 44% similar

(Rushton, 1988). Another study found that an optimal amount of

similarity is not 100%. Women prefer the scents of men with

genes somewhat similar to their own—not those of men with

genesnearlyidenticalorwithgenestotallydissimilartotheirown

(Jacob, McClintock, Zelano, & Ober, 2002). Each woman’s

choice was based on the human leukocyte antigen (HLA) gene

sequence—the basis for personal odors and olfactory prefer-

ences—inherited from her father, but not her mother. In a third

study, people rated the attractiveness of faces of the opposite sex,

including of their own faces morphed to look like a face of the

oppositesex.Bothmenandwomenratedtheirownmorphedfaces

as the most attractive, even though they did not recognize the

photos as images of themselves (Penton-Voak, Perrett, & Pierce,

1999). In yet another study, people said they trusted a stranger’s

face more when it had been morphed with their own than when it

wasunchanged(DeBruine,2002).Familiaritywasruledoutasan

explanation of the results by using morphs of celebrities; only

self-resemblance mattered.

The evidence that people tend to be similar to their spouses

and friends is pervasive (Rushton, 1989a). The correlations are

greatest on sociodemographic variables such as age, ethnicity,

and educational level (r 5 .60), next largest for opinions and

attitudes (r 5 .50), and then for IQ and cognitive ability (r 5 .40).

The correlations are smallest,but still significant, for personality

traits (r 5 .20) and physical characteristics (r 5 .20). Even

marrying across ethnic lines may ‘‘prove the rule.’’ In Hawaii,

men and women who married cross-ethnically were more similar

in personality to their spouses than were those who married

within their group, suggesting that couples ‘‘make up’’ for ethnic

dissimilarity by choosing spouses more similar to themselves in

other respects (Ahern, Cole, Johnson, & Wong, 1981).

A genetic contribution to the preference for between-spouse

similarity is implied by the finding that within sets of homoge-

neous traits, similarity is more pronounced on those more herit-

able. In the case of physical attributes, Russell, Wells, and

Address correspondence to J. Philippe Rushton, Department of Psy- chology, University of Western Ontario, London, Ontario N6A 5C2, Canada; e-mail: [email protected].

PSYCHOLOGICAL SCIENCE

Volume 16—Number 7 555Copyright r 2005 American Psychological Society at TEXAS WOMANS UNIV on June 12, 2016pss.sagepub.comDownloaded from

Rushton (1985) found spousal similarity was greater on more

heritable features such as middle-finger length (80% heritable)

than on less heritable ones such as upper-arm circumference

(50% heritable). For personality and leisure-time pursuits,

Rushton and Russell (1985) found spousal similarity greater for

more heritable items such as enjoying reading (41% heritable)

than for less heritable items such as having many different hob-

bies (20% heritable). In cognitive ability, Rushton and Nicholson

(1988) found spousal resemblance was most marked on the more

heritableof15subtestsfromthebatteryoftestsusedintheHawaii

Family Study of Cognition and 11 subtests from the Wechsler

Adult Intelligence Scale. Spouses who are similar on the most

heritable traits report greater marital satisfaction than spouses

who are similar on less heritable traits (Russell & Wells, 1991).

In a study of friends and acquaintances, Rushton (1989b)

compared long-term (nonhomosexual) male friends of European

ancestry with respect to the same 10 blood groups used in the

study of sexual partners (Rushton, 1988). Friends were signifi-

cantly more similar (54%) than random pairs drawn from the

same sample (48%). They were also more similar on a wide range

of physical, personality, and social attitude measures, and sim-

ilaritywasmorepronouncedonthemoreheritableattitudes,such

as agreement with the death penalty (51% heritable), than on the

less heritable attitudes, such as agreement with Bible truth (25%

heritable). Tesser (1993) extended these results to liking in ac-

quaintances by manipulating people’s beliefs about how similar

they were to other people on attitudes preselected as being either

highorlowinheritability.Hefoundthat peopleliked others more

when they were similar on more heritable attitudes than when

they were similar on less heritable attitudes.

Twin and adoption studies also suggest the preference for

similarity is heritable. Rowe and Osgood (1984) analyzed data

from 530 adolescent monozygotic (MZ) twins (who share 100% of

their genes) and dizygotic (DZ) twins (who share 50% of their

genes) and found antisocial behavior was about 50% heritable.

Moreover,thecorrelationof.56betweenanindividual’sdegreeof

delinquency and that of his friends was mediated genetically,

with adolescents disposed to delinquency being genetically in-

clined toseek outasfriendsotheradolescentswhowere similarly

inclined. In a study of 396 siblings from both adoptive and

nonadoptive homes, Daniels and Plomin (1985) found that

whereas the friends of biological siblings resembled each other,

those of adoptive siblings did not.

Not all twin studies have concluded that resemblance is due to

genetic preferences. Lykken and Tellegen (1993) reported a

study of 901 pairs of married twins and 1,052 of their spouses. A

number of their findings favored a genetic preference for simi-

larity: (a) Spouse pairs resembled each other on 88 out of 88

variables (personality, social attitudes, religiosity, and leisure-

time interests), (b) these variables were moderately to substan-

tially heritable, (c) the spouses of MZ twins rated their spouse’s

friends as more similar to themselves than did the spouses of DZ

twins (44% vs. 30%), and (d) the spouses of MZ twins liked their

spouse’s co-twin more than did the spouses of DZ twins (39% vs.

25%). However, Lykken and Tellegen were more impressed that

(a) difference scores between the spouses of MZ twins were no

smaller than those between the spouses of DZ twins and hardly

larger than those for random pairs (the correlation equivalents

were .14 vs. .11 vs. 0.00, respectively) and (b) MZ twins did not

liketheirco-twin’sspousemorethandidDZtwins(39%vs.37%).

The present study of twins, spouses, and friends used model-

fitting procedures to examine genetic and environmental hy-

pothesesaboutthebasisforsocialassortment.Modelfittinggives

more accurate estimates of genetic and environmental influence

than do difference scores or correlations because model fitting

takes more of the variance into account (Plomin, DeFries,

McClearn, & McGuffin, 2001). Our raw data were the variances

and covariances between and within twin pairs. The between-

pairs mean squares reflect both pair resemblances and pair

differences, and the within-pairs mean squares reflect pair dif-

ferences. We tested several non-mutually exclusive hypotheses

about why partners resemble each other.

METHOD

Questionnaires assessing demographic, attitudinal, and per-

sonality information were mailed to approximately 1,400 twin

pairs from the adult twin register at the University of London

Institute of Psychiatry. Each twin received three questionnaires,

one for him- or herself, one for the twin’s spouse, and one for the

twin’s same-sex best friend. These were marked ‘‘Twin,’’ ‘‘Same-

Sex Friend (Non-Relative),’’ and ‘‘Spouse.’’ About 480 twins

responded. One year later, a follow-up questionnaire was mailed

to those twins who had not responded but whose co-twin had,

increasing the sample size nearly 25%. Each respondent was

instructed to complete the questionnaire independently. Twins

were to return a best friend’s and a spouse’s questionnaire along

with their own in a prestamped, addressed envelope.

Each twin’s questionnaire contained 236 items; the question-

naires for the friends and spouses contained 152 items. For the

present study, we analyzed responses to the 130 items common

to twins, friends, and spouses. These items assessed physical

characteristics (height and weight; 2 items); demographic back-

ground, suchas educational level (3 items); social attitudes,such

as political affiliation (3 items); and 122 items from the Person-

ality Research Form (PRF; Jackson, 1974) and the Eysenck

Personality Questionnaire (EPQ; Eysenck & Eysenck, 1975)

measuring the scales of Autonomy and Cognitive Structure (from

the PRF) and Extraversion, Neuroticism, Psychoticism, and Lie

(from the EPQ).

Thepresentstudyisbasedonthereturnsfrom322twinpairs:174

MZ pairs (131 sister pairs, with 232 best friends and 145 spouses;

43 brother pairs, with 71 best friends and 44 spouses) and 148 DZ

pairs (82 sister pairs, with 149 best friends and 87 spouses; 28

brother pairs, with 47best friends and 16 spouses; 38opposite-sex

556 Volume 16—Number 7

Twins’ Mate Choice and Friendship

at TEXAS WOMANS UNIV on June 12, 2016pss.sagepub.comDownloaded from

pairs, with 64 best friends and 30 spouses). The returns were

complete for both friends in 140 pairs of MZ twins and 118 pairs of

DZ twins,and forboth spousesin 73pairs of MZ twinsand 50pairs

ofDZtwins.Thereturnrate,degreeofgreaterresponsefromwomen

thanmenandfromMZtwinsthanDZtwins,andrepresentativeness

of the general population are comparable to what has been found

in previous studies with this register and other volunteer twin

samples (Rushton, Fulker, Neale, Nias, & Eysenck, 1986).

RESULTS

Missing data (less than 2% of the items) were replaced using the

sex-specificitemmodefortheentiresamplecomprisingalltwins,

spouses, and friends combined (N 5 1,529). Respondents’ ages

ranged from 18 to 75 years, with a mean of 32 years. Respon-

dents came from mostly middle- and upper-middle-class family

backgrounds, had some postsecondary education, and enjoyed

an above-average level of income and residence. The means,

standarddeviations,andinternalconsistenciesofthepersonality

scales (Cronbach’s a) were similar to those in previously pub- lished studies with this register (mean a across all items 5 .73). For 488 of the twins, 8-year test-retest correlations on the four

EPQ scales ranged from .51 to .71 (ps < .01).

Table 1 reports the similarity correlations on 14 variables

for the MZ twins (r 5 .53) and DZ twins (r 5 .32), including

opposite-sexpairsandafteradjustingfortheeffectsofageandsex.

The simplest hypothesis about partner selection is that partners

will be similar. Our results are consistent with the literature in

that the age- and sex-adjusted correlations for all 322 spouse-

spouse pairs (r 5 .32), 563 friend-friend pairs (r 5 .20), and 290

spouse-friend pairs (r 5 .31) were higher for sociodemographic

and social attitude variables (mean r across all relationships 5

.38)thanforpersonalityvariables(mean r5 .17).Themeanlevel

of similarity in friends was significantly higher than zero and

significantly lower than the mean level of similarity in spouses or

in DZ twins (ps < .05). The MZ and DZ twins did not differ in how

similar they were to their respective partners. Also shown in

Table 1 are simple heritabilities calculated by doubling the dif-

ference between the MZ and DZ twins. These values are typical

of those reported in the literature (Bouchard & Loehlin, 2001;

Plomin et al., 2001).

To test whether social assortment is a heritable tendency, we

compared the MZ and DZ twins for the following cross-twin

similarities: Twin 1 and Twin 2’s spouse, Twin 1 and Twin 2’s

friend, Twin 1’s spouse and Twin 2’s spouse, Twin 1’s friend and

Twin 2’s friend, and Twin 1’s spouse and Twin 2’s friend. These

relationships were all higher for the MZ than for the DZ twins (rs

5 .17 vs. .09, .23 vs. .17, .23 vs. .14, .22 vs. .18, and .17 vs. .08,

respectively), giving simple heritabilities of 16, 12, 18, 8, and

18%. Because the MZ and DZ cross-twin multivariate genetic

analysis was limited by the phenotypic correlation of .32 for

spouses and .20 for friends (mean r 5 .26), the bivariate herita-

bility averaged 31% in spouses and 21% in friends.

We also used structural equation models to test genetic and

environmentalhypotheseswiththe14variablestakentogetherasa

latent trait of similarity. Models were fit to the age- and sex-cor-

rected variance-covariance matrices using the computer program

AMOS4.01(Arbuckle,1999). We partitioned thetotalphenotypic

variance into three sources: V(A), additive genetic effects; V(CE),

the common environment shared by a pair of twins, also known

as the between-families environmental variance; and V(E), the

unique,nonsharedenvironment,orthatportionoftheenvironment

thatisspecifictoeachtwininapair.Bydefinition,MZtwinshavea

coefficient of relatedness of 1.00 and DZ twins of .50.

TABLE 1

Similarity Correlations (and Heritabilities) for Twin Pairs, Spouses, and Friends

Correlations

Variable Monozygotic

twins (174 pairs) Dizygotic

twins (148 pairs) Spouses

(322 pairs) Friends

(563 pairs) Heritability

(%)

Height .89 .52 .21 .04 74

Weight .80 .49 .25 .08 62

Education .79 .69 .55 .56 20

Occupation .23 .24 .74 .21 0

Income .11 .18 .43 .17 0

Political views .58 .34 .60 .36 48

Altruism .38 .30 .22 .16 16

Religion .51 .42 .41 .20 18

Extraversion .51 .18 .06 .10 66

Psychoticism .50 .26 .25 .15 48

Neuroticism .56 .11 .01 .09 90

Lies .53 .16 .24 .19 74

Cognitive structure .58 .33 .29 .32 50

Autonomy .43 .33 .15 .17 20

Average .53 .32 .32 .20 42

Volume 16—Number 7 557

J. Philippe Rushton and Trudy Ann Bons

at TEXAS WOMANS UNIV on June 12, 2016pss.sagepub.comDownloaded from

Table 2 shows the results of the model-fitting analyses. For the

comparisonsofTwin1andTwin2’sspouseandTwin1andTwin2’s

friend,wefoundthatbetween13and30%ofthe variancewasdue

toadditivegeneticvariance,lessthan10%wasduetothecommon

environment, and 70 to 80% was due to each twin’s specific en-

vironment.Wefoundsimilarresultsforthespouse-spouse,friend-

friend, and spouse-friend comparisons—that is, 17 to 35% of the

variance was due to genetic effects, 10 to 30% to common envi-

ronment, and 40 to 75% to chance effects. Correcting for unreli-

ability by dividing the A and C components by Cronbach’s alpha

gave the following values for all relationships combined: A 5

34%, C 5 12%, and E 5 54%. Because the ACE and AE models

gavebetterfitsthanthosewithouttheAcomponent,weconcluded

that some of the variance was due to genetic influence. Similarly,

because the ACE and CE models gave better fits than those

without the C component, we concluded that some of the vari-

ance was due to common environmental influences. We also

tested a nonadditive (genetic dominance) model, but it did not

provide a better fit than the additive (nondominance model).

To explore further whether the preference for similarity in

social partners is genetically influenced, we examined whether

the assortment was more pronounced on the more heritable than

on the less heritable of the 130 items. We calculated the simple

heritabilities for all items making up the traits in Table 1 (fol-

lowing Neale, Rushton, & Fulker, 1986). Both spouse-spouse

(r 5 .17, p < .05) and friend-friend (r 5 .20, p < .05) similarities

were greater on more heritable items, such as ‘‘being a talkative

person’’ (50% heritable), than on less heritable items, such as

‘‘having many different hobbies’’ (20% heritable). When we ag-

gregated over both partners combined (friends and spouses), the

value of r was .22, p < .01.

DISCUSSION

The present study first replicated the well-established finding

that MZ twins (who share 100%of their genes) are nearlytwice as

similar in personality and social attitudes as DZ twins (who share

50% of their genes; see Table 1). This result suggests at least

moderate heritability for these characteristics. The study also

replicated the finding that spouses and best friends resemble

each other on these traits. The resemblance of spouses and

friends apparently falls between the resemblances of full-sib-

lings (e.g., DZ twins, as in Table 1) and half-siblings (who would

show approximately half the degree of resemblance found for DZ

twins). The magnitude of these spousal and friendship correla-

tions is higher than a previous estimate of the resemblance

between spouses and friends, which was on the order of the

correlations between cousins and half-cousins (based on studies

of assortative mating in animals; Rushton, 1989a).

ThepresentstudyalsofoundthatthefriendsandspousesofMZ

twinsweremoresimilartoeachotherthanwerethoseofDZtwins.

MZ twins chose partners more similar to those of their co-twins

than did DZ twins. Both correlational and model-fitting analyses

showed that atleast10 to30% of thevariance in partner choice is

due to genetic factors, at least 10% is due to shared environ-

mentalfactors,andtheremaining60%isduetouniqueorchance

environmental events. Correcting the figures for measurement

error raised the genetic contribution to 34% and the common

environmental component to 12% and reduced the chance fac-

tors to 54%. The genetic contribution to social assortment may

not be surprising given that MZ twins are more similar than DZ

twins in almost every preference that has been investigated, in-

cluding choice of wardrobe, job, and vacation (Bouchard &

Loehlin, 2001; Lykken & Tellegen, 1993). From arrays of pos-

sible alternatives, people seek those compatible with their geno-

types (Scarr, 1996). People prefer their own kind—extraverts

favor extraverts; traditionalists, traditionalists. That partner

similarity is most pronounced on the more heritable of items from

homogeneous arrays suggests that the preference is finely tuned.

Geneticsimilaritytheoryprovidesanevolutionaryexplanation

of why people resemble their friends and spouses. This extension

of Hamilton’s (1964) inclusive fitness theory goes beyond the kin

TABLE 2

Model-Fitting Results for the 14 Measures in Table 1

Parameter estimate (%) Chi-square goodness of fit

Comparison Number of pairs

Additive genetic

effects (A)

Common environmental effects (C)

Specific environmental effects (E)

ACE model (df 5 3)

AE model (df 5 4)

CE model (df 5 4)

E model (df 5 5)

Twin 1 and Twin 2’s spouse 322 13 7 80 40.89n 41.02 41.25 51.23

Twin 1 and Twin 2’s friend 563 30 0 70 98.16n 98.16n 105.23 129.49

Twin 1’s spouse and

Twin 2’s spouse 123 29 28 43 3.3n 4.6 5.3n 13.2

Twin 1’s friend and

Twin 2’s friend 258 35 0 65 2.7n 2.7n 8.0 22.5

Twin 1’s spouse and

Twin 2’s friend 290 17 9 74 40.2n 43.6 40.2n 66.70

nBest-fitting model determined by Dw2 (difference) test, p < .05.

558 Volume 16—Number 7

Twins’ Mate Choice and Friendship

at TEXAS WOMANS UNIV on June 12, 2016pss.sagepub.comDownloaded from

group to include spouses (who are typically not genetically re-

lated), friends, and fellow ethnics (Rushton, 1989a; Rushton,

Russell, & Wells, 1984; Salter, 2003). If you like, become friends

with, come to the aid of, and mate with those people who are

genetically most similar to yourself, you are simply trying to

ensure that your own segment of the gene pool will be safely

maintained and eventually transmitted to future generations.

Hamilton’s parameters for inclusive fitness are mainly applied to

the evolution of altruism and nepotism in immediate kin in those

species—ants, bees, or naked mole rats, for example—in which

individuals have no offspring and exist only to nurture other

relatives(Hauber&Sherman,2001).Yetifgenescanensuretheir

own survival by leading people to help family members, with

whom they share genes, then genes can also ensure their survival

by leading people to help any organism with whom they share

genes.

The model-fitting analyses foundthat some variancein partner

choice is due to the effects of common family environment.

Similarity in lifestyle preferences, including religious and po-

litical values acquired from parents, may be sought in partners.

There is evidence that human preference for facial features can

be acquired through sexual imprinting on opposite-sex parents

(Bereczkei et al., 2004). Animal studies likewise show that fa-

miliarity and proximity can modify innate preferences for indi-

viduals that smell, look, or behave like self, mother, or father

(Hauber & Sherman, 2001). Two main mechanisms of ‘‘kin rec-

ognition’’ have been identified: (a) location (‘‘if it’s in your nest,

it’s yours’’) and (b) self-referent phenotype matching (‘‘look for

physical features that are similar to self’’). Most phenotype

matching depends on a mix of inherited and learned constraints.

Up to two thirds of the variance in this study was due to unique

environmental effects. This is a residual term composed of many

sources, including measurement error and various kinds of ge-

netic and environmental interactions. It also includes chance

factors, such as being in the right place at the right time. Simi-

larity, of course, is only one of many criteria people use in

choosing social partners. Physical appearance, status, control of

resources, reciprocity, location, and family situation all provide

constraints and exert influence as well.

REFERENCES

Ahern, F.M., Cole, R.E., Johnson, R.C., & Wong, B. (1981). Personality

attributes of males and females marrying within vs. across racial/

ethnic groups. Behavior Genetics, 11, 181–194. Arbuckle, J.L. (1999). AMOS 4.01 [Computer software]. Chicago:

SmallWaters Corp.

Bereczkei, T., Gyuris, P., & Weisfeld, G.E. (2004). Sexual imprinting in

human mate choice. Proceedings of the Royal Society of London, Series B, 271, 1129–1134.

Bouchard, T.J., Jr., & Loehlin, J.C. (2001). Genes, evolution, and per-

sonality. Behavior Genetics, 31, 243–273. Daniels, D., & Plomin, R. (1985). Differential experience of siblings in

the same family. Developmental Psychology, 21, 747–760.

DeBruine,L.M.(2002).Facialresemblanceenhancestrust.Proceedings of the Royal Society of London, Series B, 269, 1307–1312.

Eysenck, H.J., & Eysenck, S.B.G. (1975). Manual for the Eysenck Per- sonality Questionnaire. San Diego, CA: Educational and Industrial Testing Service.

Hamilton, W.D. (1964). The genetical evolution of social behavior: I and

II. Journal of Theoretical Biology, 7, 1–52. Hauber, M.E., & Sherman, P.W. (2001). Self-referent phenotype match-

ing: Theoretical considerations and empirical evidence. Trends in Neurosciences, 24, 609–616.

Jackson, D.N. (1974).Personality ResearchForm manual (2nded.).Port Huron, MI: Research Psychologists Press.

Jacob, S., McClintock, M.K., Zelano, B., & Ober, C. (2002). Paternally

inherited HLA alleles are associated with women’s choice of male

odor. Nature Genetics, 30, 175–179. Lykken, D.T., & Tellegen, A. (1993). Is human mating adventitious or

the result of lawful choice? A twin study of mate selection. Journal of Personality and Social Psychology, 65, 56–68.

Neale, M.C., Rushton, J.P., & Fulker, D.W. (1986). Heritability of item

responses on the Eysenck Personality Questionnaire. Personality and Individual Differences, 7, 771–779.

Penton-Voak, I.S., Perrett, D.I., & Pierce, J. (1999). Computer graphic

studies of the role of facial similarity in judgements of attractive-

ness. Current Psychology, 18, 104–117. Plomin, R., DeFries, J.C., McClearn, G.E., & McGuffin, P. (2001). Be-

havioral genetics (4th ed.). New York: Freeman. Rowe, D.C., & Osgood, D.W. (1984). Heredity and sociological theories

of delinquency: A reconsideration. American Sociological Review, 49, 526–540.

Rushton, J.P. (1988). Genetic similarity, mate choice, and fecundity in

humans. Ethology and Sociobiology, 9, 329–333. Rushton, J.P. (1989a). Genetic similarity, human altruism, and group

selection. Behavioral and Brain Sciences, 12, 503–559. Rushton, J.P. (1989b). Genetic similarity in male friendships. Ethology

and Sociobiology, 10, 361–373. Rushton, J.P., Fulker, D.W., Neale, M.C., Nias, D.K.B., & Eysenck, H.J.

(1986). Altruism and aggression: The heritability of individual

differences. Journal of Personality and Social Psychology, 50, 1192–1198.

Rushton, J.P., & Nicholson, I.R. (1988). Genetic similarity theory, in-

telligence, and human mate choice. Ethology and Sociobiology, 9, 45–57.

Rushton, J.P., & Russell, R.J.H. (1985). Genetic similarity theory: A

replytoMealeyandnewevidence.BehaviorGenetics,15,575–582. Rushton, J.P., Russell, R.J.H., & Wells, P.A. (1984). Genetic similarity

theory: Beyond kin selection. Behavior Genetics, 14, 179–193. Russell, R.J.H., & Wells, P.A. (1991). Personality similarity and quality

of marriage. Personality and Individual Differences, 12, 406–412. Russell, R.J.H., Wells, P.A., & Rushton, J.P. (1985). Evidence for

genetic similarity detection in human marriage. Ethology and So- ciobiology, 6, 183–187.

Salter, F.K. (2003). On genetic interests: Family, ethny and humanity in an age of mass migration. Frankfurt, Germany: Peter Lang.

Scarr, S. (1996). How people make their own environments: Implica-

tions for parents and policy makers. Psychology, Public Policy, and Law, 2, 204–228.

Tesser, A. (1993). The importance of heritability in psychological re-

search: The case of attitudes. Psychological Review, 93, 129–142.

(RECEIVED 5/7/04; REVISION ACCEPTED 6/16/04)

Volume 16—Number 7 559

J. Philippe Rushton and Trudy Ann Bons

at TEXAS WOMANS UNIV on June 12, 2016pss.sagepub.comDownloaded from