Narcissism Guides Mate Selection: Humans mate assortatively, as revealed by facial resemblance, following an algorithm of "self seeking like

Evolutionary Psychology

human-nature.com/ep – 2004. 2: 177-194

¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯

Original Article

Narcissism guides mate selection: Humans mate assortatively, as
revealed by facial resemblance, following an algorithm of “self seeking
like”

Liliana Alvarez, Universidad Simón Bolívar, Caracas, Venezuela. Email: lalvarez@usb.ve.

Klaus Jaffe, Universidad Simón Bolívar, Apartado 89000, Caracas 1080A, Venezuela. Email:
kjaffe@usb.ve. (Corresponding author).

Abstract:

Theoretical studies suggest that mating and pair formation is not likely to

be random. Computer simulations suggested that sex among genetically complex
organisms requires mate choice strategies for its evolutionary maintenance, to reduce
excessive genetic variance produced by out-crossing. One strategy achieving this aim
efficiently in computer simulations is assortative mating modeled as “self seeking
like”. Another one is selection of “good genes”. Assortative mating increases the
probability of finding a genetically similar mate, without fomenting inbreeding,
achieving assortative mating without hindering the working of other mate selection
strategies which aim to maximize the search for “good genes”, optimizing the
working of sex in evolutionary terms. Here we present indirect evidence that in a
significant proportion of human reproductive couples, the partners show much higher
facial resemblances than can be expected by random pair formation, or as the
outcome of “matching for attractiveness” or the outcome of competition for the most
attractive partner accessible, as had been previously assumed. The data presented is
compatible with the hypothesis derived from computer simulations, that human mate
selection strategies achieve various aims: “self seeking like” (including matching for
attractiveness) and mating with the best available genes.

Keywords

: mate selection, face recognition, assortative mating, sex, evolution

¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯

Introduction

What is the adaptive value of love? We certainly do not know, but we might

get closer in answering this question by understanding the evolutionary mechanisms
underlying mate choice. Although many theoretical studies assume mating to be
random, recent computer simulations showed that random mating is very unlikely to

Narcissism Guides Mate Selection

occur in nature (Kalick and Hamilton 1986, Jaffe

1996, 1998). Specifically,

theoretical studies have suggested that assortative mating seems to be highly adaptive
(Thiessen and Gregg 1980, Davis 1995), as it reduces excessive allelic variance
induced by recombination and sex, especially among diploids with a large genome
(Jaffe 1998, 1999, 2000). Assortative mating defined as “self seeking like” has a
strong stabilizing effect on sex, is evolutionary stable, and has an evolutionary
dynamics analogous to kin selection (Jaffe 2000). In addition, assortative mating
affects the genetic structure of populations, influencing the evolutionary dynamics of
sexual organisms significantly (Dieckmann and Doebeli

1999, Kondrashov and

Kondrashov 1999, but see Ochoa and Jaffe 1999) and thus, is a feature that should be
taken into account when studying the adaptive value of behaviors related to mate
selection. The theoretical works mentioned suggest that assortative mating in itself is
beneficial in evolutionary terms to the organism practicing it, and thus is likely to be
widespread in nature.

The interpretation of evidence for assortative mating is controversial (Moore

1992). Data can be interpreted in the light of incest avoidance mechanisms or in that
of optimizing outbreeding. Living organisms seem to optimize rather than maximize
outbreeding (Bateson 1983). That is, mate choice mechanisms avoid maximizing
outbreeding and inbreeding at the same time. A complementary theory to an incest-
avoidance-outbreeding equilibrium is the optimization of the working of sex (Jaffe
1999, 2000, 2002). This theory accepts that genetic similarity is not only achieved
through familiar proximity, and recognizes that genetic relatedness may exist among
individuals with no familiar relationship between them. Therefore, assortative mating
of the kind “self seeking like” may achieve reproduction between genetically similar
mates, favoring the stabilization of genes supporting social behavior, with no kin
relationship among them (Jaffe 2001).

Revising the experimental evidence for assortative pairing at the molecular

level, Tregenza and Wedell (2000) found evidence that suggests that genetic
compatibility limits mate choice. Recent studies with lizards (Dickinson and Koenig
2003, Sinervo and Clobert 2003) showed that after removing all spatial surrogates for
kinship, lizards still settle and actively chose to cooperate with phenotypically and
genetically similar lizards.

Evidence for assortative mating among humans seems well established.

Human's mate assortatively regarding age, IQ, height, weight, nationality, educational
and occupational level, physical and personality characters (Buston and Emlen 2003,
Buss 1989, Epstein and Guttman 1984, Garrison et al.

1968, Ho 1986, Jaffe and

Chacon 1995, Spuhler

1968), and family relatedness (Rushton 1989, Spuhler 1968,

but see Genin et al.

2000, Isles et al. 2001). Even Homer in his Odyssey (XVII, 218)

wrote that “god always joins those who are similar”. Yet, assortative mating is
evidently limited by very well known mechanisms of inbreeding avoidance among
humans (see for example reviews in van den Berghe 1983, Wolf 1993).

Humans place much weight on the visual aspect of faces. Leonardo Da Vinci

(1452-1519) wrote in his notebook, when referring to how to select beautiful faces to

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 178 -

Narcissism Guides Mate Selection

paint, that the artist should search for faces regarded as beautiful by the public rather
than by himself, as his own wit might deceive him as it will lead him to look for faces
similar to himself. Thus, if he has an ugly face, he will paint ugly faces, unless he
searches for the public taste (Da Vinci 1999).

The human face is a complex, unique and characteristic pattern, most familiar

to us when distinguishing people (Vezjak and Stephancic 1994). For example,
studies have shown that people remember faces of their own race better than faces of
other races; that the recognition memory for same-race is superior to other-race faces;
and that differential activation in fusiform regions contributes to same-race memory
superiority (Golby et al. 2001). Hauber and Sherman

(2001) showed how

mechanisms of self reference have a neurophysiological basis. Faces seem to be
involved in reproductive behavior among humans. Couples faces resemble each other
much more than random pair formation would suggest (Griffiths and Kunz 1973,
Zajonc et al 1987, Hinsz 1989). Similarities between faces are not likely to arise as a
result of pair formation or environmental factors (Rice and Borecki 2001), as facial
features have a strong genetic basis (Savoye et al. 1998). Facial resemblance between
couples has been extensively reviewed recently (Penton-Voak and Perrett 2000) and
can certainly be viewed as an adaptive trait, product of evolutionary forces and not an
experimental artefact. Human faces are even considered as a communication device
for the advertisement of some kind of heritable quality (Thornhill and Gangestad
1999) and thus should provide reliable signals in the search for mates with “good
genes” (Jaffe 1999).

Imprinting, i.e. memorizing in early age the visual images of parents and then

using these images for mate choice, as first discovered in birds (Lorenz 1935), also
seems to guide assortative mating in humans (Todd and Miller1993, Penton-Voak
and Perrett

2000, Bereczkei et al 2002, Perrett et al 2002, Little et al.

2003). Other

evidence, pointing to the existence of parts of the mechanism needed to allow humans
“imprint” the faces of their parents, was provided by Le Grand et al.

(2001). They

showed the need of “early” visual input to develop normal face recognitions later.
Children resemble their parents (Nesse et al 1990, Bredart and French 1999, McLain
et al 2000, Oda 2002, but see Bressan and Grassi 2004), sometimes even in odd ways:
they seem first to resemble more their fathers (see also Daly and Wilson 1982,
Regalski and Gaulin 1993). Facial child-parent resemblance mechanisms seem to
exist even among chimpanzee (Parr and de Waal 1999). This visual memory may
then be use to establish criteria for beauty, which in turn are used to select a mate,
producing as a consequence assortative mating. These and other evolutionary effects
of parental imprinting have been discussed by Todd and Miller (1993).

Yet how is this assortment achieved among humans? What are the behavioral

and psychological processes that achieve assortment of couples in human society?
Mainly three theories are known to address this question: The matching hypothesis;
assortative mating based on “self seeks like”; and the competition for the most
attractive mate which achieves assortative mating as a kind of Nash equilibrium
outcome.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 179 -

Narcissism Guides Mate Selection

The “

self seeking like” hypothesis

, assumes a multidimensional space for

individual preferences so that every “self” is unique. Data on assortative pairing
based on facial visual cues, favoring the “self seeking like” hypothesis, include the
finding by DeBruine (2002) that facial resemblance enhances trust. That is, inborn
psychological mechanisms originally evolved for kin selection and mate selection
seem to serve as a basis of other more advanced developments of social behavior.
One such behavior in humans could well be the outcome of cognitive processes
underlying human mate choice, where self-perception seems to modulate mate
preference (Buston and Emlen 2003). Evidence that humans look for a self, or that
they root criteria of beauty on the self, are also compatible with this hypothesis
(Aron. and Aron 1986, Yela and Sangrador 2002).

More commonly accepted, though, is the

competition hypothesis

, where the

physical and psychological resemblance between mates is thought to be the outcome
of a competition for the most attractive partner. Some evidence supports this
assumption as similar degrees of attractiveness have also been found between
partners in reproductive couples (Berscheid and Walster 1974, Murstaein and Christy
1976). These and other authors do not regard similar attractiveness between partners
of a couple as a unequivocal sign of assortative mating (Kalick and Hamilton 1986)
but more likely as the outcome of competition, where more attractive individuals will
mate with the most attractive partner available to him or her (Miller and Todd 1998
for example). Other variables are known to also affect attractiveness and thus should
influence pair formation. These could be adaptive for basic physiological reasons
such as those suggested by Penton-Voak et al. (1999) who showed that menstrual
cycle alters face preference.

Under the competition hypothesis, assortative mating is the outcome of a

mechanism based on general levels of attractiveness. That is, highly attractive
females have a first choice for highly attractive males, and/or vice versa (Miller and
Todd 1998). Attractiveness seems to be a rather general and broad attribute applied to
different stimuli, visual or not. A specific piece of evidence for this was recently
provided by Collins and Missing (2003) who showed that vocal and visual
attractiveness are related in women.

A variant of the competition hypothesis is provided by the

matching

hypothesis,

which

proposes that we don’t seek the most physically attractive person

but that we are attracted to individuals who match us in terms of physical attraction
(Kalick and Hamilton 1986). This compromise is thought to be necessary because of
a fear of rejection (a more attractive person might reject your advances) and/or to
achieve a balance between partners (Walster et al. 1966). Murstein (1972) obtained
indirect support for the matching hypothesis. The physical attractiveness of engaged
couples and those going out together was judged from photographs. There was a
definite tendency for the two people in each couple to be similar in terms of physical
attractiveness (Murstein and

Christy1976).

One way to discriminate between the various theories is to recognize that

attractiveness is a one-dimensional space whereas similarity is a multidimensional

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 180 -

Narcissism Guides Mate Selection

one. That is, if a universal sense of beauty creates a basis for a universal scale of
attractiveness, then pairing would proceed from the most attractive extreme
downwards in an assortative fashion: the best pair up with the best, then the second-
best pair up, and so on. On the other hand, similarity has no universal extremes. If
assortative mating is based on similarities (other or in addition to attractiveness), then
paring would not form a hierarchical linear scale of attractiveness.

Thus, if theoretical predictions about assortative mating are correct, and

physical features of faces are largely determined by genetic factors, we should detect
assortative mating based on facial visual cues. If assortative pairing is the outcome of
competition for the most attractive partner, partners in a couple should have similar
levels of attractiveness, and little heterogeneities in level of attractiveness between
partners of couples should be expected. Here we present evidence that assortative
mating based on facial visual cues occurs in human populations, and that these facial
similarities seem to be the product of “self seeking like” rather than of the
competition for the most attractive accessible partner.

Methods

We photographed 36 randomly selected couples, from a list of addresses

provided by a local doctor in the city of Mérida, Venezuela, which either had children
and/or were living for at least 3 years together, and which reported to have no known
family relationship between them. A digital black and white photograph of each of
the partners was taken in or around their homes (Figure 1). The subjects of these 36
couples were not used for any of the behavioral tests performed and we will call them
the target subjects.

To assess the existence of resemblance between the faces of couples among

the target subjects, the photographs of the males were placed on a table and those of
the females were randomly shuffled. The test subjects (over 100 volunteers at the
universities in Caracas and Mérida) had to assign each of the photographs of female
target subjects to one of the males. Test subjects did not know any of the target
subjects photographed. The test was performed double blind, as neither the
experimenter nor the test subject knew who the correspondence of the photos to the
real couples.

The amount of correct guesses, i.e., joining photographs of male and female

partners of the same couple, did not differ between test subjects asked to "Choose the
female to which the male is most likely to be married" and test subjects asked to
"Choose the female that is most likely to be a sibling to the male" (p > 0.55, Chi
squared test, df = 19. The questions given in quotations are rough translations from
Spanish). Thus, for further tests, we asked the test subjects to "Join the photographs
of the putative couples with the closest resemblance and or the more likely to be
married".

In order to simplify the tests and reduce the rejection of test subjects to

participate, we built 6 pools of 6 couples (i.e. 12 target subjects each), so that in each

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 181 -

Narcissism Guides Mate Selection

To test for the effect of cues in the background of the photo in identifying

correctly couples from photographs of target subjects, we prepared 3 experiments.
We manipulated the digital photos so as to provide: 1- the same background for each
pair in a couple but which differed between couples; 2- a different background for
each of the partners in a couple; 3- no background for all couples which made all
photos to have the same background (as in Figure 1). As an additional test to reject a
possible effect of background cues in the photographs, we also presented photos of
only the mouth, eyes or nose of the subjects (see Figure 1). The test subjects were
then asked to pair the photos of the facial features of target subjects according to their
similarity.

As face recognition abilities are dependent on early visual experience (Le

Grand et al.

2001) and are better between individuals of the same race (Golby et al.

2001), we correlated the number of correct guesses made by test subjects with the
skin colour difference between the test subject and the average of the faces of the
correctly guessed couples. The colour was assessed by the experimenter for both test
and target subjects in 3 distinct categories (dark, slightly dark, white), which were
assigned values of 1 to 3 in order to perform statistical correlations on the results. Not
all test subjects were used for this test as only skin colour data assessed by one of the
experimenters was used

In order to estimate the effect of attractiveness of faces as a possible

confounding factor, we assessed the attractiveness of a face, through test subjects
from the opposite sex. The attractiveness was registered from less attractive to very
attractive in a scale from 1 to 4. Twenty one subjects of each sex were presented with
the photographs of 35 members of couples of the opposite sex of the test subject for
this assessment. That is, test males were given the photos of target females and test
females those of target males to assess the attractiveness. Data was ordered based on
increased attractiveness of either males or females.

The statistical analyses performed on the data were applied on the scores of

test subjects. That is on the number of correct couples guessed by the test subjects.
The analyses were: Pearson correlation coefficient to assess correlations between age
and scoring, and between attractiveness and scoring. Chi square test to compare the
total number of scores obtained for a given experimental setting with those expected
for random guessing. The tests involved that each test subject had to match all photos
for all coupes. Random guessing under this scenario for either 36 pairs or 6 pairs
gives in average one correct guess per test subject. Deviation from one was assumed
to be non-random guessing and its significance was assessed by the Chi-squared test.
The degrees of freedom were calculated as the number of test subjects used for that
given experimental setting, minus one. Another more sensitive way to look at the
results was to assess the number of times a given couple was correctly identified as
such by test subjects. This distribution of guesses (see Figure 2) was then compared
with an expected distribution obtained by random guessing. The outcome of random
pair formation plus random guessing was estimated using a simple Monte Carlo
simulation model written in basic.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 183 -

Narcissism Guides Mate Selection

Figure 2

: Number of times a couple was correctly guessed by test subjects compared

to the expected frequency of correct guesses from a sample of 36 couples, assessed
with a Monte Carlo simulation, assuming random pair formation and random
guessing (line with small dots).

Results

The number of correct guesses, i.e. guessed pairs of photographs

corresponding to actual couples, made by tests subjects was far larger than expected
by random guessing in most experiments. When females were provided with the
photos of the target faces of 36 couples, they guessed correctly an average of 2.5
couples (Significantly different from random, n = 25 test females, p < 0.0001, Chi-
square = 132). Male test subjects placed in front of the same task managed to identify
correctly only an average of 0.94 couples (Not different from random, n= 18 test
males, p=0.6, Chi-square = 15). The amount of correct guesses made by female test
subjects was significantly higher than those made by male test subjects (p < 0.003,
chi-square = 38).

When the test was simplified, so that only the photos of faces of 6 couples

were presented at the same time, this difference between the number of correctly
guessed couples achieved by female and male test subjects disappeared (p = 0.11,

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 184 -

Narcissism Guides Mate Selection

Chi-square = 21). The average number of correctly guessed couples was 1.71 and
1.91 for female and male test subjects respectively (Significantly different from
random in both cases, n = 35 and 21, p < 0.0001 in both cases).

No significant correlation (

= 0.1) between correct guesses and age of the

test subjects was found (neither for females r = 0.009, mean age = 27, range 19-52,
nor for males r = -0.2, mean age = 25, range = 18-70).

We correlated the number of correct guesses made by test subjects with the

skin colour difference (dark, slightly dark, white) between the test subject and the
average of the faces of the correctly guessed couples, and obtained a Pearson’s
correlation coefficient r = 0.5 (p < 0.01, n = 65). Thus, if test subjects were given
same race faces to evaluate, the number of correct guesses increased.

Table 1:

Scores, assessed as number of correctly guesses target couples, produced by

test subjects when confronted with photographs of faces from partners from 36
couples.

Number of correct guesses

0 1 2 3 4 5 Total

Males

7 6 4 1 0 0 18

Females

4 5 3 4 5 4 25

Total

11 11 7 5 5 4 43

When examining the distribution on the number of correct guesses received

by each target couple we found that many test subjects never guessed correctly any
couple (Table 1). That is, 39 % of males and 16 % of females never scored a single
correct guess, and 28 % of males and 64 % of females scored above one (random). Of
the 36 target couples, 64 % were guessed more than once (chance) and 44 % more
than two times (Fourth column of Table 4). We compared the number of times each
couple was guessed correctly with that predicted by a Monte Carlo simulation based
on random pair formation and random guessing. These comparisons, for all the
experiments performed, are given in Table 2, where we compare the distribution of
correct guesses with that produced by Monte Carlo simulations, using a chi-squared
test . The table reads as follows. In the first line, for example, we present the results
of tests using photographs of faces having all the same background. The number of
photos provided to the test subjects was 72 (36 photos of females and 36 of males,
belonging to 36 couples). The test subjects for this experiment were both male and

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 185 -

Narcissism Guides Mate Selection

females and 43 test subjects were tested. The chi-square value obtained when
comparing the actual data with that from a Monte Carlo simulation with totally
random guessing, as explained in Figure 2, was 352. This value indicates that the
actual data differs from totally random guessing significantly, as the odds of
obtaining the actual data by chance are <<0.0001 for a distribution function of
guesses with 13 degrees of freedom (see Figure 2).

Table 2

: Comparison between a distribution of guesses achieved randomly (Monte

Carlo simulations) and the distribution of the number of correct assignments made by
test subjects, guessing the partners of couples, based on photographs of faces or parts
of faces.

Photos of

Nr of

pairs

Test subjects

doing the

guessing

Nr of

tests

df p

All couples same background

Both sexes

352 13 <<0.0001

All couples same background

Females

112 10

<0.001

All couples same background

Males

= 0.001

Couples with same background

but different between couples

6 Females 9

<0.001

Couples with same background

but different between couples

6 Males 9

<0.001

Couples with same background

but different between couples

6 Both

sexes 18 259

<<0.0001

Couples with different

background

6 Females 9

140

<<0.0001

Couples with different

background

6 Males 7

115

<<0.0001

Couples with different

background

6 Both

sexes 16

>1000

<<0.0001

Mouths of couples

Both sexes

>1000 14 <<0.0001

Noses of couples

Both sexes

>1000 15 <<0.0001

Eyes of couples

Both sexes

>1000 18 <<0.0001

The results presented in Table 2 show that in all cases studied test subjects

were able to correctly identify a significant proportion of couples with an accuracy
that far exceeded that expected by random guessing or by random pair formation.
With this more sensitive test, even male test subject which were much less accurate
compared to females in guessing the correct pairs for the couples if the sample of
photographs to choose from was large, showed a score that statistically differed from
random guessing.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 186 -

Narcissism Guides Mate Selection

The results in Table 2 also show that the background of the photographs did

not affect significantly the correct guessing of pairs. Independently of how the
background of the photos was presented, test subjects could pick out the real couples
based on facial similarities. Test subjects were even able to correctly guess the real
couples if photos of only parts of the faces of target subjects (nose, eyes or mouth)
were presented. This recognition was less accurate than if the complete face was
presented, but the large amount of tests performed provide for a high statistical
significance of this effect.

When ordering couples based on the attractiveness of the faces of one of the

sexes, the average value of attractiveness of the opposite sex correlated strongly with
the attractiveness of the sex used to order the couples. That is, more attractive females
tended to pair with more attractive males, or vice-versa. This tendency can be
visualized in Table 3. This table shows that the average attractiveness, as assessed by
20 test subjects of the opposite sex, was lowest for female mates of the most
unattractive males and highest for the female mates of the most attractive males,
when the couples were ordered according to the attractiveness of the male partner.
When the data was order according to the attractiveness of the female partner in the
couple, an even stronger result was obtained. The average attractiveness of male
mates of the least attractive females was much lower than that of male mates of the
most attractive females.

Table 3:

Mean attractiveness of the partner of opposite sex, in couples in a given

quartile if ordered by attractiveness of:

Males Females

Quartile 1 (Less attractive) 1.78

1.35

Quartile 2

1.76

1.73

Quartile 3

2.08

2.03

Quartile 4 (Very attractive) 2.13

2.56

Table 4:

Scores obtained by each couple (number of correct guesses), the average

attractiveness index obtained by each individual, and the difference in attractiveness
between the partners of the couple (absolute difference) as assessed by neutral judges.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 187 -

Narcissism Guides Mate Selection

Couple

Attractiveness

male

Attractiveness

female

Score:

No. of correct

guesses

Difference

attractiveness

2 2 0

0.00

2.11 2.09 8 0.02

2.33 2.25 4 0.08

2 2.09 16

0.09

1.8 1.64 0 0.16

1.83 1.64 0 0.19

1.5 1.3 4 0.20

1.7 1.5 2 0.20

2.22 2.42 2 0.20

1.5 1.27 2 0.23

1.1 1.33 12 0.23

1.75 2 0 0.25

2.1 1.8 0 0.30

1.5 1.8 4 0.30

1.7 2 4

0.30

1.22 1.56 6 0.34

1.6 2 4

0.40

2.2 1.79 0 0.41

1.8 1.38 0 0.42

1.5 1.93 0 0.43

1.9 1.44 6 0.46

1.13 1.6 0 0.47

1.7 1.22 2 0.48

2.33 1.77 8 0.56

2.1 2.67 2 0.57

2.22 2.81 2 0.59

2 2.67 4 0.67

1.67 2.42 0 0.75

1.33 2.18 18 0.85

2.44 1.56 8 0.88

3.3 2.36 14 0.94

1.89 2.92 4 1.03

2.2 1 2

1.20

3.38 2.08 0 1.30

1.4 3 0

1.60

Discussion

Our results confirm that human couples resemble each other significantly

more than expected for random pair formation, and that this resemblance can be
detected by neutral judges (test subjects). In the sample used here, a significant
proportion of the couples studied showed to have conspicuous facial similarities,

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 188 -

Narcissism Guides Mate Selection

detectable by judges. This result confirms that our test was sufficiently sensitive and
the sample examined was adequate, regarding the number of couples with
conspicuous facial similarities between them, for the assessment of assortative
mating.

We wanted to know if this assortment was the product of “self seeking like”

or the outcome of competition for the most attractive partner available. Assortment
based on attractiveness was certainly at work in our sample as shown with our
experiments assessing the attractiveness of the partners of the couples, presented in
Table 3. That is, there was a general tendency for more attractive males to pair with
more attractive females and vice versa. Yet, attractiveness by itself could not explain
our results for various reasons:

1- Test subjects correctly guessed the real couples if photos of only parts of

the faces of target subjects (nose, eyes or mouth) were presented. Even though the
criteria for attractiveness are not applicable or are much more difficult to apply on
only parts of faces, these parts revealed sufficient information on similarities for test
subjects to re-assemble photographs of real life couples.

2- The lack of correlation between the scores of correct guesses and the

difference in attractiveness between partners in the couples (Table 4) is not
compatible with the competition hypothesis or the matching hypothesis. If assortative
mating was the outcome of competition for attractiveness, “similarity” and matching
for attractiveness should correlate positively. In addition, if matching attractiveness is
at work, the assessment of similarity by test subjects should also be based on levels of
attractiveness. This was certainly not the case in our study as the scores for similarity
and the differences in levels of attractiveness between the partners of a couple
diverged conspicuously. Thus, the competition hypothesis by itself can not explain
our results.

3- Criteria for attractiveness vary between races and thus, an absolute scale of

attractiveness for humans is not likely to exist. This fact has been known for quite a
wile as discussed in the introduction. Our data confirmed that the ability to recognize
similarities in faces was related to race, making it unlikely that assortative mating can
be explained solely on the basis of matching universal attractiveness criteria or as the
outcome of competition for the most attractive mate available.

The cumulative evidence presented here favors the hypothesis that humans

search for couples based on “self seeks like”, using a narcissistic psychological
algorithm in assessing the appropriate mate. Yet passive assortative mating could also
explain our results. Passive assortative mating occurs when, due to population
viscosity, reproduction occurs among spatially proximate individuals that are
probably close or distant relatives. In the active sort, individuals choose their mates
based on similar phenotypic traits, which reflect similar genes. In both cases the result
is assortative mating or breeding among mates that possess similar genes.

Our results showed that females are better than males in assessing facial

resemblance between individuals when a large number of choices are presented to

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 189 -

Narcissism Guides Mate Selection

test subjects. This phenomenon is congruent with the fact that females have a much
finer discriminatory ability than males (Briceño and Jaffe1997).

The results of this study are compatible with the notion that humans develop a

sense of beauty through imprinting-like mechanisms. This sense of beauty must have
a strong narcissistic component, as it is formed through the images of the parents, as
discussed in the introduction. When this sense of beauty is applied to mate selection,
the outcome is assortative mating in a multidimensional scale, as no universal scale of
beauty can be formed through this mechanism. Our results can not discard that
assortative mating in humans is at least partially achieved through competition for the
most attractive potential partner, or by matching attractiveness. Yet, our results
strongly suggest that a multitude of other visual criteria are involved in mate
selection. The hypothesis of “self seeking like” explains the experimental results
rather well. The fact that these narcissistic criteria seem to be applied also in situation
were no pairs for reproductive purposes are involved, such as in the choice of partners
for business purposes (DeBruine 2002), strongly support the narcissist hypothesis. A
testable prediction to possibly falsify the ‘self seeking like” hypothesis is that
narcissistic criteria should be applied to many other situations in human every day life
involving aesthetic or affective assessments. For example the choice of pets should
also follow narcissistic criteria. This has been shown to be the case after obtaining our
results (Roy and Christenfeld 2004, Payne and Jaffe 2004).

In the present study we seem to have dealt just with one envelop of a rather

complex set of mate selection mechanisms. Thus, further research should refine the
mechanisms at work that seem to balance assortative mating with inbreeding
avoidance and selection for ‘good genes’ (Jaffe 2002). The results might direct future
research in focusing on the still unknown intricacies of mate selection mechanisms
and the origin of love.

Acknowledgements:

We thank Omar Arenas for advice in the statistical treatment of

the data, Peter Todd, John Hutchinson and Diana Ajami for helpful comments,
Oswaldo Henriquez and Rodolfo Jaffe for help in recruiting test subjects. L. A.
received financial support from CONICIT.

Received 8 May, 2004, Revision received 10 November, 2004, Accepted 19
September, 2004.

References

Aron, A. and Aron, E. N. (1986).

Love and the Expansion of Self: Understanding

Attraction and Satisfaction

. New York: Hemisphere Publish Corporation.

Bateson, P. (1983). Optimal outbreeding. In Bateson, P. (Ed.)

Mate Choice

(pp. 257-

277). Cambridge: Cambridge University Press.

Bereczkei, T., Gyuris, P., Koves, P., and Bernath, L. (2002). Homogamy, genetic

similarity, and imprinting; parental influence on mate choice preferences.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 190 -

Narcissism Guides Mate Selection

Personality and Individual Differences

, 33: 677-690.

Berscheid, E. and Walster, E. (1974). Physical attractiveness. In Berkowitz, L. (Ed.),

Advances in Experimental Social Psychology, (

pp.

157-215). New York:

Academic Press.

Bredart, S. and French, R.M. (1999). Do babies resemble their fathers more than their

mothers? A failure to replicate Christenfeld and Hill (1995).

Evolution and

Human Behavior

, 20: 129–135.

Bressan, P. and Grassi, M. (2004). Parental resemblance in one-year-olds and

the Gaussian curve.

Evolution and Human Behavior

, 25: 133-141

Briceño, A. and Jaffe, K. (1997). Sex differences in occupational performance: a

biological perspective.

Social Biology

, 44: 198-204.

Buss, D. (1989). Sex differences in human mate preferences: Evolutionary

hypotheses tested in 37 cultures

. Brain and Behavioral Sciences,

14: 519-520.

Buston, P. M. and Emlen, S. T. (2003). Cognitive processes underlying human mate

choice: The relationship between self-perception and mate preference in
Western society. www.pnas.orgycgiydoiy10.1073ypnas.1533220100

Collins, S. A. and Missing, C. (2003). Vocal and visual attractiveness are related in

women.

Animal Behaviour

, 2003, 65: 997-1004

Da Vinci, L. (1999). Cuadernos de notas. Traducción: J. L. Velaz. Edimat Libros,

Madrid, page 134.

Daly, M. and Wilson, M. (1982). Whom are newborn babies said to resemble?

Ethology and Sociobiology,

3: 69-78

Davis, C. H. (1995). The effect of assortative mating and environmental variation on

selection for sexual reproduction.

Evolutionary Theory

, 11: 51-53.

DeBruine, L.M. (2002). Facial resemblance enhances trust.

Proceedings of the Royal

Society London B

, 269: 1307-1312

Dickinson, J.L. and Koenig, W. D. (2003). Desperately seeking similarity.

Science

300: 1887-1889.

Dieckmann, U. and Doebeli, M. (1999). On the origin of species by sympatric

speciation.

Nature,

400: 354-357.

Epstein, E. and Guttman, R. (1984). Mate selection in man: Evidence, theory and

outcome.

Social Biology

, 31: 243-278.

Garrison, R., Anderson, E. and Reeds, S. (1968). Assortative marriage.

Social

Biology

, 15: 113-127.

Genin, E., Ober, C., Weitkamp, L. and Thomson, G. (2000). A robust test for

assortative mating.

European Journal of Human Genetics

, 8: 119-124.

Golby, A.J., Gabrieli, J. D. E., Chiao, J. Y. and Eberhardt, J. L. (2001). Differential

responses in the fusiform region to same-race and other-race faces.

Nature

Neuroscience,

4: 845-850.

Griffiths, R. W., and Kunz, P. R. (1973). Assortative mating: A study of

physiognomic homogamy.

Social Biology

, 20: 448-453.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 191 -

Narcissism Guides Mate Selection

Hauber, M.E. and Sherman, P.W. (2001). Self-referent phenotype matching:

theoretical considerations and empirical evidence.

Trends Neuroscience,

24:

609-616

Hinsz, V. B. (1989). Facial resemblance in engaged and married couples.

Journal of

Social and Personal Relationships

, 6: 223-229.

Ho, H. S. (1986). Assortative mating in unwed-birth parents? Adoptive and

nonadoptive parents.

Social Biology,

33: 77-86.

Isles, A. R,, Baum, M. J., Ma, D., Keverne, E. B. and Allen, N. D. (2001). Genetic

imprinting: Urinary odour preference in mice.

Nature

, 409: 783-784.

Jaffe, K. and Chacon, G. (1995). Assortative mating: Sex differences in mate

selection for married and unmarried couples.

Human Biology

, 67: 111-120.

Jaffe, K. (1996). On the dynamics of the evolution of sex or why the sexes are, in

fact, always two?

Interciencia

, 21: 259-267 and 22: 48 (erratum).

Jaffe, K. (1998). Sex, mate selection and evolution. In Porto, V. W., Saravanan, N.,

Waagen, D. and Eiben, A. E. (Eds.)

Lecture Notes in Computer Science 1447:

Evolutionary Programming VII

(pp. 483-492). Springer Verlag.

Jaffe, K. (1999). On the adaptive value of some mate selection strategies.

Acta

Biotheoretica

, 47: 29-40.

Jaffe, K. (2000). Emergence and maintenance of sex among diploid organisms aided

by assortative mating.

Acta Biotheoretica,

48: 137-147.

Jaffe, K. (2001). On the relative importance of Haplo-Diploidy, Assortative Mating

and Social Synergy on the Evolutionary Emergence of Social Behavior.

Acta

Biotheoretica

, 49: 29-42.

Jaffe, K. (2002). On sex, mate selection and evolution: an exploration.

Comments on

Theoretical Biology

, 7: 91-107.

Kalick, S. M., Hamilton, T. E. (1986). The matching hypothesis reexamined.

Journal

of Personality and Social Psychology,

51: 673-682.

Kondrashov, A. S. and Kondrashov, F. A. (1999). Interactions among quantitative

traits in the course of sympatric speciation.

Nature

, 400: 351-354.

Le Grand, R., Mondloch, C. J., Maurer, D. and Brent, H. P. (2001). Early visual

experience and face processing.

Nature,

410: 890.

Little, A. C., Penton-Voak, I. S., Burt, D. M. and Perrett, D. I. (2003). Investigating

an imprinting-like phenomenon in humans Partners and opposite-sex parents
have similar hair and eye colour.

Evolution and Human Behavior

, 24: 43–51.

Lorenz, K. (1935). Der kumpan in der Umwelt des Vogels.

Journal of Ornithology

83: 137-213.

McLain, D. K., Setters, D., Moulton, M. P. and Pratt, A. E. (2000). Ascription of

resemblance of newborns by parents and nonrelatives.

Evolution and Human

Behavior

, 21: 11–23.

Miller, G. F. and Todd, P. M. (1993). Evolutionary wanderlust: Sexual selection with

directional mate preferences. In Meyer, J.-A., Roitblat, H. L. and Wilson, S.
W. (Eds.).

From Animals to Animats 2: Proceedings of the Second

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 192 -

Narcissism Guides Mate Selection

International Conference on Simulation of Adaptive Behavior

(pp. 21-30),

M.I.T. Press, Bradford Books.

Miller, G. F. and Todd, P. M. (1998). Mate choice turns cognitive.

Trends in

Cognitive Sciences

, 2: 190-198

Moore, J. (1992). Sociobiology and incest avoidance: a critical look at a critical

review.

American Anthropologist,

94: 930-933

Murstein, B. I. (1972). Physical attractiveness and marital choice.

Journal of

Personality and Social Psychology

, 22: 8–12.

Murstein, B. I., and Christy, P. (1976). Physical attractiveness and marriage

adjustment in middle-aged couples.

Journal of Personality and Social

Psychology

, 34: 537- 542.

Nesse, R. M., Silverman, A. and Bortz, A. (1990). Sex differences in ability to

recognize family resemblance.

Ethology and Sociobiology

, 11: 11–21.

Ochoa, G and Jaffe, K. (1999). On sex, mate selection and the Red Queen.

Journal of

Theoretical Biology

, 199: 1-9.

Oda, R., Matsumoto-Oda, A. and Kurashima, A. (2002). Facial resemblance of

Japanese children to their parents.

Journal of Ethology

, 20: 81–85.

Parr, L. and de Waal, F. (1999). Visual kin recognition in chimpanzees.

Nature,

399:

647-648.

Payne C. and Jaffe, K. (2004). Self seeks like: Many humans choose their dog-pets

following rules used for assortative mating.

Journal of Ethology

(in press).

DOI: 10.1007/s10164-004-0122-6

Penton-Voak, I and Perrett, D. I. (2000). Consistency and individual differences in

facial attractiveness judgements: An evolutionary perspective.

Social

Research

, 67: 219-245.

Penton-Voak I., Perrett D., Castles D., Kobayashi T., Burt D., Murray L. and

Minamisawa R. (1999). Menstrual cycle alters face preference.

Nature

, 399:

741-742.

Regalski, J. M. and Gaulin, S. J. C. (1993). Whom are Mexican infants said to

resemble? Monitoring and fostering paternal confidence in the Yucatan.

Ethology and Sociobiology,

14: 97-113.

Rice, T. K. and Borecki, I. B. (2001). Familial resemblance and heritability.

Advances

in Genetics,

42: 35-44.

Roy M. M. and Christenfeld N. J. S. (2004).

Do Dogs Resemble Their Owners?

Psychological Science

, 15: 361-363

Rushton, J. P. (1989). Genetic similarity, human altruism and group selection.

Brain

and Behavioral Sciences

, 12: 503-559.

Savoye, I., Loos, R., Carels, C., Derom, C. and Vlietinck, R. (1998). A genetic study

of anteroposterior and vertical facial proportions using model-fitting.

The

Angle Orthodontist,

68: 467-470.

Sinervo, B. and Clobert, J. (2003). Morphs, dispersal behavior, genetic similarity and

the evolution of cooperation.

Science

, 300: 1949-1851.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 193 -

Narcissism Guides Mate Selection

Spuhler, J. (1968). Assortative mating with respect to physical characteristics.

Social

Biology

, 15: 128-140.

Thiessen, D. and Gregg, B. (1980). Human assortative mating and genetic

equilibrium: An evolutionary perspective.

Ethology and Sociobiology

, 1: 111-

140.

Thornhill, R. and Gangestad, S. W. (1999). Facial attractiveness.

Trends in Cognitive

Sciences

, 3: 452-460

Todd, P. M. and Miller, G. F. (1993). Parental guidance suggested: How parental

imprinting evolves through sexual selection as an adaptive learning
mechanism.

Adaptive Behavior

, 2: 5-47.

Tregenza, T. and Wedell, N. (2000). Genetic compatibility, mate choice and patterns

of parentage: invited review.

Molecular Ecology,

9: 1013-27

van den Berghe, P. (1983). Human inbreeding avoidance: Culture in nature.

Behavioral and Brain Sciences,

6: 91-123.

Vezjak, M. and Stephancic, M. (1994). An anthropological model for automatic

recognition of the male human face.

Annals Human Biology

21: 363-380

Walster, E., Traupmann, J. and Walster, G.W. (1978). Equity and extramarital

sexuality.

Archives of Sexual Behavior.,

7: 127-41.

Wolf, A. P. (1993). “Westermarck Redivivus”.

Annual Review Anthropology,

22:

157-175.

Yela, C. and Sangrador, J. L. (2002). Perception of physical attractiveness throughout

loving relationships.

Current Research in Social Psychology,

6: 5

Zajonc, R., Adelmann, P., Murphy, S. and Niendenthal, P. (1987). Convergence in

the physical appearance of spouses.

Motivation and Emotion

, 11: 335-346.

Evolutionary Psychology – ISSN 1474-7049 – Volume 2. 2004.

- 194 -

Document Outline