Content uploaded by Tamás Bereczkei
Author content
All content in this area was uploaded by Tamás Bereczkei on Feb 19, 2014
Content may be subject to copyright.
Received 14 November 2003
Accepted 17 December 2003
Published online 29 April 2004
Sexual imprinting in human mate choice
Tamas Bereczkei
1
, Petra Gyuris
1
and Glenn E. Weisfeld
2*
1
Department of Psychology, University of Pe
´
cs, Ifjusag utja 6, H-7624, Hungary
2
Department of Psychology, Wayne State University, Detroit, MI 48207, USA
Animal and human studies have shown that individuals choose mates partly on the basis of similarity, a
tendency referred to as homogamy. Several authors have suggested that a specific innate recognition
mechanism, phenotypic matching, allows the organism to detect similar others by their resemblance to
itself. However, several objections have been raised to this theory on both empirical and theoretical
grounds. Here, we report that homogamy in humans is attained partly by sexual imprinting on the
opposite-sex parent during childhood. We hypothesized that children fashion a mental model of their
opposite-sex parent’s phenotype that is used as a template for acquiring mates. To disentangle the effects
of phenotypic matching and sexual imprinting, adopted daughters and their rearing families were exam-
ined. Judges found significant resemblance on facial traits between daughter’s husband and her adoptive
father. Furthermore, this effect may be modified by the quality of the father–daughter relationship during
childhood. Daughters who received more emotional support from their adoptive father were more likely to
choose mates similar to the father than those whose father provided a less positive emotional atmosphere.
Keywords: evolved mating preferences; homogamy; sexual imprinting; phenotype matching
1. INTRODUCTION
Research has shown that human sex partners resemble
each other in many traits (Mascie-Taylor 1988, 1995;
Weisfeld et al. 1992; Jaffe & Chacon-Puignau 1995; Keller
et al. 1996; Bereczkei et al. 1997; Thiessen 1999). Positive
correlations have been found between their socio-
economic status, age, intellectual ability, education,
personality variables, physical attractiveness, vocational
interests and anthropometric measures. Although hom-
ogamy, or positive assortative mating, appears to be under
the influence of many cultural factors that presumably
render married couples more compatible, its ubiquity
across cultures and species requires an evolutionary expla-
nation. One possible explanation is genetic-similarity
theory, an extension of kin-selection theory (Thiessen &
Gregg 1980; Rushton 1989). If organisms can identify
genetic similarity in strangers, they might exhibit altruism
towards them, with a consequent increase in inclusive fit-
ness. This altruism could take the form of homogamy,
which increases the degree to which parents share genes
with offspring, thereby enhancing genetic representation
in future generations. Homogamy might also prevent gen-
etic complexes coadapted to the local environment from
being disrupted, thereby enhancing reproductive success
(Read & Harvey 1988). Obviously, an extreme degree of
homogamy can impose serious reproductive costs
(inbreeding depression effects) (Blouin & Blouin 1988). It
has been argued that an adaptive compromise has evolved
between the opposing selection pressures of inbreeding
and outbreeding, with individuals choosing a mate with a
moderate degree of homogamy (Bateson 1983). Accord-
ingly, homogamy between non-relatives has been found to
enhance marital stability and fertility, and to lower the rate
*
Author for correspondence (weisfeld@sun.science.wayne.edu).
Proc. R. Soc. Lond. B (2004) 271, 1129–1134 1129 2004 The Royal Society
DOI 10.1098/rspb.2003.2672
of miscarriage in humans (Mascie-Taylor 1988; Rushton
1988; Thiessen et al. 1997).
How is homogamy thought to be achieved? Some ani-
mals have been found to prefer genetically similar mates
(Blaustein et al. 1991; Holmes 1995; Pfennig & Sherman
1995). This may occur by phenotypic matching to self,
through which they use their own phenotype to choose
others with similar phenotypic cues. Obviously, this can
occur only if there is a high correlation between genetic
similarity and phenotypic similarity on traits that individ-
uals use to distinguish potential mates. Much experi-
mental evidence shows that various animals are able to
recognize genetic similarity in unfamiliar individuals on
the basis of shared olfactory and visual cues. For example,
a study of Belding’s ground squirrels (Spermophilus
beldingi) revealed that biological sisters reared apart could
recognize one another and engaged in a significantly lower
rate of aggressive interactions than non-siblings
(Holmes & Sherman 1983).
However, very few studies have examined the possibility
that humans recognize gene-related ‘strangers’ based on
similarity to their own phenotype. These studies have
yielded controversial results. One has shown that a woman
prefers the odour of a man who has significantly more
human leucocyte antigen (HLA) allele matches to her own
alleles than the man with the least preferred odour (Jacob
et al. 2002). Another study revealed that female students
tended to prefer the scent of men who possessed dissimilar
HLA genotypes (Wedekind & Furi 1997). It is not known
what perceptual mechanisms are involved in this feature
detection, and what part of the similarity detection system
is based on innate mechanisms and associative learning.
Furthermore, critics argue that the theory of genetic simi-
larity via phenotypic matching has a possible theoretical
fallacy (Archer 1989; Krebs 1989; Daly et al. 1997). As
relatedness decreases, the probability that individuals
share the gene responsible for altruism will be lower, and
distant relatives or strangers are not likely to carry that
1130 T. Bereczkei and others Sexual imprinting in human mate choice
gene, even though they may be similar in certain traits.
Therefore, discriminatory altruism on the basis of pheno-
typic similarity is uncorrelated with the likelihood of shar-
ing an altruism gene, and would be selected out. As a
consequence, genetic similarity theory will be reduced to
kin recognition.
Imprinting-like mechanisms have been suggested as an
alternative explanation for the mediation of homogamy in
humans (Daly 1989; Bereczkei et al. 2002; Little et al.
2002). Similarity between spouses would arise if individ-
uals acquired mate-choice criteria templates through
exposure to their parents. In cases of sexual imprinting,
as the ethologists called it, an early fixation to a set of
family characteristics will shape mate preferences during
adulthood (Bateson 1964; Lorenz 1965; Bolhuis & Horn
1992). Cross-fostering experiments with birds and mam-
mals have revealed that adults prefer sexual partners that
are similar to the opposite-sex parent that reared them
(Immelmann et al. 1991; Oetting et al. 1995; Kendrick et
al. 1998). For example, supporting his optimal outbreed-
ing theory, Bateson (1988) found that Japanese quail
(Corturnix japonica) prefer to mate with birds that differ
slightly in appearance from their parents and siblings.
Sexual imprinting on the observed features of the
opposite-sex parent during a sensitive period in early
childhood might be responsible for shaping people’s later
mate-choice criteria. Children might internalize their
opposite-sex parent’s phenotype as a template and, at sex-
ual maturity, prefer those who resemble this mental
model. In a recent study comparing more than 300 facial
photographs of family members and controls, the subjects
correctly matched wives to their husband’s mother
(Bereczkei et al. 2002). Furthermore, a higher degree of
similarity was perceived between the husband’s mother
and his wife than between the husband and his wife. These
results suggest that sons fashion a representation of their
mother’s physical appearance and use it for choosing
mates. In another study, similarity was observed between
the facial photographs of husbands and their wife’s bio-
logical father (Gyuris 2003). The subjects correctly
matched husbands to their father-in-law at a significantly
higher mean rate than expected by chance, and a signifi-
cantly higher frequency than matching between husbands
and wives.
However, a crucial limitation of these investigations is
the difficulty of separating the effects of phenotype match-
ing to self and sexual imprinting. Similarity between one’s
spouse and his/her opposite-sex parent might be an arte-
fact, given the 50% overlap between the parents’ and off-
spring’s genetic material. Therefore, if homogamy works
via phenotype matching, it would be responsible for the
similarity between spouse and opposite-sex parent. In this
case, our results about resemblances between family mem-
bers may be a result of innate similarity detection between
spouses, not sexual imprinting on the mother. However,
certain studies suggest that imprinting-like mechanisms
may occur in human mate choice. One study revealed
greater importance of the opposite-sex parent over the
same-sex parent in predicting the hair and eye colour of
actual partners (Little et al. 2002). Similarly, women were
found to prefer the odours of men with HLA alleles that
resembled her father’s HLA alleles but not her mother’s
(Jacob et al. 2002).
Proc. R. Soc. Lond. B (2004)
To disentangle these effects further, the present study
aimed at investigating the mate choices of women from
adoptive families. In light of our hypothesis, sexual
imprinting will have a long-term effect on one’s mate
choice, whether the child and the caring adult were rela-
tives or not. We predicted, therefore, that the women
would choose a mate who resembled their father even
though he was not a biological relative. Alternatively, if
phenotypic matching theory is correct, the women would
prefer mates similar to themselves but not to their adop-
tive father.
2. METHOD
(a) Subjects
To test this hypothesis, we applied a method that was used
in an earlier study (Bereczkei et al. 2002). Subjects were shown
photographs of wives, and rated the similarity between each of
them and four possible husbands, one of whom was the actual
spouse. They also rated the degree of similarity between the
wife’s adoptive father at the age when his adopted daughter was
between 2 and 8 years, and her possible husbands. Similarly,
facial photographs of the wife and her adoptive mother were
compared. These photographs were provided by 26 Caucasian
families. Advertisements were placed in three newspapers in
Pecs (Hungary), stating that researchers at the University of
Pecs needed volunteers who had been reared in an adoptive fam-
ily. Additional photographs, as controls, were taken of 198
young men at the age comparable with that of the husbands
(mean age of 28.3 years). They were randomly chosen from
undergraduate and graduate students at university. We took
black-and-white photographs because many of the mothers’ and
fathers’ portraits had been taken several decades ago in this for-
mat. The photographs of the adoptive parents had been taken
when they were young, i.e. when their adopted children were
growing up. All of the photographs were developed and scanned
into a computer, using the Corell Photopaint program.
Three sets of tableaux were made of these individual photo-
graphs. They were standardized for age, length and colour of
hair, position of their head in the photograph, and certain other
traits, to minimize the possible differences between the faces of
target persons and controls. One tableau contained the wife (left
side), and her husband and three controls (right side). The other
set of tableaux was made of the same photographs, with the
exception that the wife was replaced by her adoptive father
(figure 1). The third set showed the adoptive mother next to the
same group photographs (one husband, three controls). Finally,
we had three sets of tableaux, with a total of 182 photographs
of individual faces.
In all, 242 undergraduate (128 female, 114 male) psychology
students were used as independent judges to evaluate similarity.
They were not aware of the aim of the study, and were not per-
sonally acquainted with any of the persons in the photographs.
Eighty-four subjects participated in the first study (wife–hus-
band matching), 82 in the second study (husband–father) and
76 in the third study (husband–mother).
(b) Procedure
Each tableau was projected on a screen in a seminar room,
where up to 10 subjects participated in the experiment at a time.
The subjects were instructed to look at each of the 3 × 26 tab-
leaux thoroughly, match the photographs of the index person
with four possible matches, one of which was the true match.
Sexual imprinting in human mate choice T. Bereczkei and others 1131
(a)
(b)(c)
(d )(e)
Figure 1. The tableau shows adoptive father when his daughter was between 2 and 8 years old (a) and four possible sons-in-
law (b)–(e). Subjects were asked to make an assessment about the rank of similarity between them. The appropriate match
was (d ).
The subjects ranked the pairs individuals on the basis of simi-
larity. They marked the degree of similarity with numbers from
1 (the most similar) to 4 (the least similar). In the first three
studies, they judged the similarity between the husband and
wife, between the husband and his wife’s adoptive father, and
between the husband and his wife’s adoptive mother, respect-
ively.
In the final study, wives were asked to complete a retrospec-
tive attachment test, the EMBU, which assesses adults’ percep-
tions of their parents’ rearing behaviour during childhood. The
short form of the EMBU, with 23 items, provides three fac-
torially derived subscale measures: rejection (e.g. ‘my parents
criticized me and told me how lazy and useless I was in front of
others’); emotional warmth (e.g. ‘my parents showed with words
and gestures that they liked me’); and overprotection (e.g. ‘my
parents wanted to decide how I should be dressed or how I
should look’). In an examination among samples of 2442 stu-
dents from four countries, internal consistency and reliability has
been high (Arindell et al. 1999). In our study, the test was
administered in the subject’s home by a research assistant who
was naive about the point of the study. For each EMBU item,
participants rated themselves on a scale ranging from 1 (no,
never) to 4 (yes, most of the time). Finally, the perceived degree
of similarity between husband and his father-in-law was corre-
lated with the test scores on the EMBU.
3. RESULTS
(a) Study 1: husband–wife matching
As figure 2 shows, the resemblance between husbands
and their wives exceeded chance. The judges correctly
matched husbands to wives at a significantly higher mean
rate than 25% (31.10%, Wilcoxon z = ⫺5.86, p ⬍ 0.01).
This percentage significantly exceeds the rate at which a
control was chosen as most similar (22.9%, z = ⫺6.124,
p ⬍ 0.001). No sex difference was found in the estimation
of similarity; approximately the same proportion of male
and female subjects made the appropriate matches
(Mann–Whitney U = 234.3, p ⬎ 0.05).
Proc. R. Soc. Lond. B (2004)
40
35
30
25
20
husband–wife husband–father husband–mother
percentage
Figure 2. The percentage of matches of husband to wife,
husband to wife’s adoptive father, and husband to wife’s
adoptive mother. Significant resemblance was ascribed
between husband and wife, and between husband and wife’s
adoptive father, with a significant difference between the two
mean rates.
(b) Study 2: husband–father matching
Even more similarity was perceived between husbands
and their wives’ adoptive father (figure 2). The subjects
correctly chose the husband as being the most similar to
his father-in-law, on average, in 37.7% of the tableaux,
which exceeds chance level (z = ⫺7.81, p ⬍ 0.001). Hus-
bands were ranked first on the similarity scale almost twice
as often as controls (20.7%, z = ⫺7.618, p ⬍ 0.001). No
sex difference was found between subjects (U = 324.3,
p ⬎ 0.05).
Comparing the results of studies 1 and 2, a higher
degree of similarity was perceived between the wife’s hus-
band and the wife’s adoptive father than between the hus-
band and his wife. The subjects correctly matched
husbands to their fathers-in-law at a significantly higher
frequency than to their wives (37.7% versus 31.1%,
z = ⫺2.85, p ⬍ 0.01).
(c) Study 3: husband–mother matching
The percentage of matches of husband to wife’s mother
was not significantly above chance level (27.2%,
1132 T. Bereczkei and others Sexual imprinting in human mate choice
14
12
10
8
6
4
0.5 1.0 1.5 2.0 2.5 3.0
percentage
scores of the adoptive father’s emotional warmth
Figure 3. Percentage of husband–wife’s adoptive father
matching as a function of the father’s emotional warmth.
The more emotional warmth the adoptive daughters received
during their childhood, the more similarity was judged
between their husband and adoptive father.
z = ⫺0.86, p ⬎ 0.5). No resemblance was found in facial
appearance between husbands and wives’ adoptive mother.
(d) Study 4: exposure to adoptive father
The degree of perceived similarity between the husband
and the wife’s adoptive father was plotted against the
mean scores on the three subscales of the EMBU (figure
3). A regression of emotional warmth on matching fre-
quencies by judges showed a significant positive relation-
ship: women who had been more emotionally supported
by their fathers during childhood were more likely to
choose mates similar to their fathers on facial traits
(

= 0.510, t = 2.90, p ⬍ 0.01). No significant relationship
was found for the other subscales: rejection (

= ⫺0.288,
t = ⫺1.285, p ⬎ 0.05) and overprotection (

= ⫺0.128,
t = ⫺0.633, p ⬎ 0.05).
This result suggests that those fathers who had provided
emotional warmth for their adopted daughters during
childhood were judged as being more similar to their
daughters’ husbands than those fathers who did not pro-
vide a positive atmosphere. In other words, those fathers
who were most frequent matches to their sons-in-law had
shown more emotional warmth towards their adopted
daughters.
4. DISCUSSION
Our results showed that: (i) significant resemblance was
found between husbands and their wives; (ii) an even
higher degree of similarity was perceived between husband
and the wife’s father (husband’s father-in-law); and (iii)
the more emotional warmth the father provided for his
adopted daughter, the more similarity was perceived
between him and his son-in-law. These findings suggest
that sexual imprinting-like mechanisms play a role in
human mate choice. Although homogamy via phenotypic
matching to self could not be excluded, homogamy may
be achieved more by exposure to the opposite-sex parent
early in life. Preferences for ensuring positive assortative
mating seem to be shaped during the process of bonding
to the opposite-sex parent, and long-term mates will be
selected partly on the basis of resemblance to that parent.
Proc. R. Soc. Lond. B (2004)
Our results could also be explained as a result of a fam-
iliarization effect; people generally respond positively to
familiar stimuli. In this case, mother and father alike
should influence their offspring’s homogamous mating,
and people would be attracted to faces with some charac-
teristics of both parents. However, because no resem-
blance was revealed between husband and the adoptive
same-sex parent, that rival hypothesis was rejected.
The recent study and our several earlier researches on
sexual imprinting (Bereczkei et al. 2002; Gyuris 2003)
found significant facial resemblance between wife and
husband. These results are not inconsistent with the possi-
bility that homogamy is promoted by phenotype match-
ing. Phenotype matching seems to influence kin
recognition. Studies have revealed that various vertebrates
can recognize relatives that were reared apart from them
(Blaustein et al. 1991; Pfennig & Sherman 1995). An
innate capacity for kin recognition would be especially
adaptive for species that disperse when young (Weisfeld
et al. 2003). In this case, an assessment of the degree of
kinship would promote altruism and cooperative behav-
iour towards individuals with shared genes.
There is some evidence that people are able to recognize
relatives who share similar facial or olfactory features. A
study found that mothers who had limited contact with
their newborns immediately after birth could recognize
them by olfactory cues alone (Porter 1987). Adult judges
could also match the odours of mothers with those of their
infants, but were unable to match husbands with wives
(Porter et al. 1985). Using visual cues, adult subjects
could match mothers’, fathers’, and their newborn infants’
facial photographs (Christenfeld & Hill 1995). Recently,
Weisfeld et al. (2003) found that participants were able to
identify the odour of most of their first-degree relatives,
but mothers could not recognize their stepchildren, nor
could children recognize their stepsiblings.
We suggest that phenotype matching has less influence
on interpersonal relationships beyond the circle of kinship.
A genetically canalized learning process (sexual
imprinting), rather than direct genetic similarity detection
via phenotype matching, is responsible for the similarity
between spouses. Our studies found that similarity
between oneself and one’s spouse was significantly
exceeded by the similarity between one’s spouse and one’s
opposite-sex parent. A recent study has revealed that eye
and hair colour were positively correlated between part-
ners but the best predictor of the partner’s eye and hair
colour was the opposite-sex parents’ colour traits (Little
et al. 2002).
Our results support the notion of a long-lasting effect
of attachment during childhood on later mating prefer-
ences. In the recent study a positive relationship was found
between the degree of facial similarity between the daught-
ers’ husband and their father and the scores of emotional
warmth provided by these fathers. Similarly, a previous
study revealed a negative correlation between maternal
rejection towards son and spouse similarity (Bereczkei et
al. 2002). All of these results suggest that mate choice
depends on physical and emotional exposure to the
opposite-sex parent, as the sexual imprinting model pre-
dicts. In accordance with this theory, individuals shape a
mental model of their opposite-sex parent’s appearance,
and search for a partner who possesses certain traits
Sexual imprinting in human mate choice T. Bereczkei and others 1133
similar to that perceptual schema. An important difference
between imprinting-like mechanisms and phenotype
matching is that the development of the former definitely
needs social interaction. In this theoretical framework,
homogamy is shaped not by a genetically prescribed rec-
ognition of similarity, but during a learning process that
occurs in a specific direction that is advantageous to gen-
etic reproduction.
One of the referees suggested that emotional investment
by the father to the adopted daughter may be mediated
by their facial resemblance. Fathers may invest more in
adopted daughters who resemble them and so daughters
selecting partners who resemble themselves would also
result in a link between the father–husband facial simi-
larity. A recent study has shown that males react more
favourably to children’s faces that have been morphed to
resemble their own (Platek et al. 2002). These are the chil-
dren whom males are most likely to adopt, and with whom
they would like to spend the most time. If the father’s
investment on adoptive daughters (including emotional
closeness) depends on the degree of their facial resem-
blance, it would explain the similarity between daughters
and husbands which would further support the
imprinting-like effect on mate choice. In a future study,
this effect could be tested by evaluating similarity between
the fathers’ and adopted daughters’ facial photographs.
At present, however, we do not know the particular
attachment and learning mechanisms that are responsible
for homogamy. It is possible that duration of co-residence
or the amount of physical contact would influence the
developmental processes through which individuals
acquire mate-choice criteria from exposure to their par-
ents. A longer period of co-residence might strengthen the
impact of sexual imprinting. However, this hypothesis was
not supported, in that we did not find a significant
relationship between the age at which daughters were
adopted and the degree of similarity between daughter’s
husband and daughter’s adoptive father. It is highly prob-
able that a complex array of environmental factors and
multiple social contacts with parents shapes a mental
model of potential partners. Recent research has shown
that certain contextual family variables, such as presence
or absence of father, trustworthiness of family members,
intensity of emotional stress and security of attachment,
tend to change later life-history traits, including the onset
of maturation, the age of first sexual experience, the length
of stable pair-bonding (Belsky et al. 1991; Kim & Smith
1998; Bereczkei & Csanaky 2001). However, it is not yet
well understood how these adaptive trajectories are linked
to the effects of the opposite-sex parent. Much more
research is needed based on the perspectives of life-history
strategy and sexual imprinting to provide a coherent pic-
ture of these developmental phenomena.
REFERENCES
Archer, J. 1989 Why help friends when you can help sisters
and brothers? Behav. Brain Sci. 12, 519–520.
Arindell, W. A. (and 11 others) 1999 The development of a
short form of the EMBU: its appraisal with students in
Greece, Guatemala, Hungary and Italy. Pers. Individ. Diff.
27, 613–628.
Proc. R. Soc. Lond. B (2004)
Bateson, P. P. G. 1964 An effect of imprinting on the percep-
tual development of domestic chicks. Nature 202, 421–422.
Bateson, P. P. G. 1983 Optimal outbreeding. In Mate choice
(ed. P. P. G. Bateson), pp. 257–277. Cambridge Univer-
sity Press.
Bateson, P. P. G. 1988 Preferences for close relations in
Japanese quail. In Acta XIX. Congress Int. Ornith., vol. 1 (ed.
H. Quellett), pp. 961–972. University of Ottawa Press.
Belsky, J., Steinberg, J. & Draper, P. 1991 Childhood experi-
ence, interpersonal development, and reproductive strategy:
an evolutionary strategy. Child Dev. 62, 647–670.
Bereczkei, T. & Csanaky, A. 2001 Stressful family environ-
ment, mortality, and child socialization: life-history stra-
tegies among adolescents and adults from unfavourable
social circumstances. Int. J. Behav. Dev. 25, 501–508.
Bereczkei, T., Vo
¨
ro
¨
s, A., Ga
´
l, A. & Berna
´
th, L. 1997
Resources, attractiveness, family commitment: reproductive
decisions in mate choice. Ethology 103, 681–699.
Bereczkei, T., Gyuris, P., Koves, P. & Bernath, L. 2002 Hom-
ogamy, genetic similarity, and imprinting; parental influence
on mate choice preferences. Pers. Individ. Diff. 33, 677–690.
Blaustein, A. R., Bekoff, M., Byers, J. A. & Daniels, T. J. 1991
Kin recognition in vertebrates: what do we really know about
adaptive value? Anim. Behav. 41, 1079–1083.
Blouin, S. F. & Blouin, M. 1988 Inbreeding avoidance
behaviors. Trends Ecol. Evol. 3, 230–233.
Bolhuis, J. J. & Horn, G. 1992 Generalization of learned pref-
erences in filial imprinting. Anim. Behav. 44, 185–187.
Christenfeld, N. J. S. & Hill, E. A. 1995 Whose baby are you?
Nature 378, 669.
Daly, M. 1989 On distinguishing evolved adaptation from epi-
phenomena. Behav. Brain Sci. 12, 520.
Daly, M., Salmon, C. & Wilson, M. 1997 Kinship: the concep-
tual hole in psychological studies of social cognition and
close relationships. In Evolutionary social psychology (ed. J. A.
Simpson & D. T. Kenrick), pp. 265–296. Lawrence
Erlbaum Associates.
Gyuris, P. 2003 Homogamy, imprinting, evolution: mate
choice of women and men rearing in various family environ-
ments. PhD thesis, University of Pe
´
cs, Hungary. [In Hung-
arian.]
Holmes, W. G. 1995 The ontogeny of littermate preferences
in juvenile golden-mantled ground squirrels: effects of rear-
ing and relatedness. Anim. Behav. 50, 309–322.
Holmes, W. G. & Sherman, P. W. 1983 Kin recognition in ani-
mals. Am. Sci. 71, 46–55.
Immelmann, K., Pro
¨
ve, R., Lassek, R. & Bischof, H. 1991
Influence of adult courtship experience on the development
of sexual preferences in zebra finch males. Anim. Behav. 42,
83–89.
Jacob, S., McClintock, M. K., Zelano, B. & Ober, C. 2002
Paternally inherited HLA alleles are associated with
women’s choice of male odor. Nature Genet. 30, 175–179.
Jaffe, K. & Chacon-Puignau, G. 1995 Assortative mating: sex
differences in mate selection for married and unmarried
couples. Hum. Biol. 67, 111.
Keller, M. C., Thiessen, D. & Young, R. K. 1996 Mate assort-
ment in dating and married couples. Pers. Individ. Diff. 21,
217–221.
Kendrick, K. M., Hinton, M. R. & Atkins, K. 1998 Mothers
may irreversibly determine male social and sexual prefer-
ences. Nature 395, 229–230.
Kim, K. & Smith, P. K. 1998 Retrospective survey of parental
marital relations and child development. Int. J. Behav. Dev.
22, 729–751.
Krebs, D. 1989 Detecting genetic similarity without detecting
genetic similarity. Behav. Brain Sci. 12, 533–534.
1134 T. Bereczkei and others Sexual imprinting in human mate choice
Little, A. C., Penton-Voak, I. S., Burt, D. M. & Perrett, D. I.
2002 Investigating an imprinting-like phenomenon in
humans. Partners and opposite-sex parents have similar hair
and eye colour. Evol. Hum. Behav. 24, 43–51.
Lorenz, K. 1965 Evolution and modification of behavior. Univer-
sity of Chicago Press.
Mascie-Taylor, C. G. N. 1988 Assortative mating for psycho-
matric characters. In Human mating patterns (ed. C. G. N.
Mascie-Taylor & A. J. Boyce), pp. 61–82. Cambridge Uni-
versity Press.
Mascie-Taylor, C. G. N. 1995 Human assortative mating: evi-
dence and genetic implications. In Human populations: diver-
sity and adaptations (ed. A. J. Boyce & V. Reynolds), pp. 86–
105. Oxford University Press.
Oetting, S., Pro
¨
ve, E. & Bischof, H. 1995 Sexual imprinting
as a two-stage process: mechanisms of information strorage
and stabilization. Anim. Behav. 50, 393–403.
Pfennig, D. W. & Sherman, P. W. 1995 Kin recognition. Sci.
Am. 272, 68–73.
Platek, S. M., Burch, R. L., Panyavin, I. S., Wasserman,
B. H. & Gallup, G. G. 2002 Reactions to children’s faces.
Resemblance affects males more than females. Evol. Hum.
Behav. 23, 159–166.
Porter, R. H. 1987 Kin recognition: functions and mediating
mechanisms. In Sociobiology and psychology: ideas, issues, and
applications (ed. C. Crawford, M. Smith & D. Krebs), pp.
175–204. London: Lawrence Erlbaum Associates.
Porter, R. H., Cernoch, J. M. & Balogh, R. D. 1985 Odor sig-
natures and kin recognition. Physiol. Behav. 34, 445–448.
Proc. R. Soc. Lond. B (2004)
Read, A. F. & Harvey, P. H. 1988 Genetic relatedness and the
evolution of animal mating patterns. In Human mating pat-
terns (ed. C. G. N. Mascie-Taylor & A. J. Boyce), pp. 115–
131. Cambridge University Press.
Rushton, J. P. 1988 Genetic similarity, mate choice, and fec-
undity in humans. Ethol. Sociobiol. 9, 329–333.
Rushton, J. P. 1989 Genetic similarity, mate choice, and group
selection. Behav. Brain Sci. 12, 503–518.
Thiessen, D. 1999 Social influences on human assortative mat-
ing. In The descent of mind. Psychological perspectives on homi-
nid evolution (ed. M. C. Corballis & S. G. Lea), pp. 311–
323. Oxford University Press.
Thiessen, D. & Gregg, B. 1980 Human assortative mating and
genetic equilibrium: an evolutionary perspective. Ethol.
Sociobiol. 1, 111–140.
Thiessen, D., Young, R. K. & Delgado, M. 1997 Social
pressures for assortative mating. Pers. Individ. Diff. 22,
157–164.
Wedekind, C. & Furi, S. 1997 Body odour preferences in men
and women: do they aim for specific MHC combinations
for simply heterozygosity? Proc. R. Soc. Lond. B 264, 1471–
1479. (DOI 10.1098/rspb.1997.0204.)
Weisfeld, G. E., Russell, R. J. H., Weisfeld, C. C. & Wells,
P. A. 1992 Correlates of satisfaction in British marriages.
Ethol. Sociobiol. 13, 125–145.
Weisfeld, G. E., Czilli, T., Phillips, K. A., Gall, J. A. & Licht-
man, C. M. 2003 Posible olfaction-based mechanisms in
human kin recognition and inbreeding avoidance. J. Exp.
Child Psychol. 85, 279–295.
