AD ALTA
JOURNAL OF INTERDISCIPLINARY RESEARCH
OBJECT-SUBJECT SIMILARITY IN FACIAL ATTRACTIVENESS ASSESSMENT (FOCUSING ON
EXTRAVERSION AND INTELLIGENCE)
a
SLÁVKA DÉMUTHOVÁ,
b
ANDREJ DÉMUTH
a
University of Ss. Cyril and Methodius in Trnava, Slovakia,
b
Trnava University in Trnava, Slovakia,
email:
a
slavka.demuthova@ucm.sk,
b
andrej.demuth@truni.sk
This work was supported by the Slovak Research and Development Agency under the
contract No. APVV-15-0294.
Abstract: According to the theory of assortative mating, the similarity/attractiveness
hypothesis and the theory of cognitive averaging, the attractiveness of a human face
can be based on the degree of similarity with the personality features of the subject
(beholder). The study tests this tendency with the example of two characteristics –
extraversion and intelligence. The subjects (N=1,903) were male (N=754; 39.6%) and
female Slovaks between 15 and 67 years of age. The T-test revealed there was no
difference in the levels of self-reported extraversion between group which preferred a
face high in extraversion and group which preferred a face low in extraversion. Similar
results were gained for intelligence variable.
Keywords: Face, attractiveness, intelligence, extraversion, assortative mating.
1 Introduction
In research into human facial attractiveness, a variety of
approaches can be seen. Besides the opinion that attractiveness is
very subjective, multi-causally influenced and a variable aspect
of the objects perceived (commonly represented by phrases such
as: “De gustibus non disputandum est” or “Beauty is in the eye
of beholder”), there are also a large number of cases that point to
the fact that there is some consensus in standards of beauty
evaluation. The vast majority of the research into the
attractiveness of the human face reports that attractive faces are
those that present with symmetry (see e.g., Fink, Neave,
Manning, & Grammer, 2006; Jones, DeBruine, & Little, 2007;
Little, Apicella, & Marlowe, 2007), averageness (Apicella,
Little, & Marlowe, 2007; Komori, Kawamura, & Ishihara,
2009), apparent sexually dimorphic traits (Perrett et al. 1998;
Burriss, Welling, & Puts 2011), etc. The explanation for the
attractiveness of these features lies in their close connection with
the genetic quality of the subject (Scheib, Gangestad, &
Thornhill, 1999), their age (Burt & Perrett, 1997), overall health
(Etcoff 1999), or fertility (Gray & Boothroyd, 2012). From this
point of view, facial features serve as an honest signal for the
potential of
‟good genes” in a perceived subject (Little, Jones,
DeBruine, & Feinberg, 2008).
Another tendency observed in this area of research is
a preference for faces that resemble the face of the observer.
This has been proved in various extrinsic facial features such as:
eye colour (see e.g., DeBruine, Jones, & Little, 2017), the shape
of the eyes, nose, mouth, or chin (Wong, Wong, Lui, & Wong,
2018), or by facial adiposity (Fisher et al., 2014). However,
within the evaluation of facial attractiveness, humans do not rely
on purely extrinsic characteristics. They are also influenced by
other factors, such as apparent social status (Buss, 1989),
intelligence (Kazanawa, 2011; Démuth & Démuthová 2018), or
personality features (Little, Burt, & Perrett, 2006) visible in the
face.
The tendency to prefer partners similar to themselves has been
explained by various theories. Firstly, the theory of assortative
mating assumes that individuals have a tendency to mate with
those who are similar to them in some way, to a higher degree
than would be expected at random (Escorial & Martín-Buro
2012). Research undertaken in this area showed a significant
degree of similarity between couples in various features of their
personalities – e.g., in agreeableness, openness to experience
(McCrae et al., 2008), cooperativeness, generosity (Tognetti,
Berticat, Raymond, & Faurie, 2014), attachment avoidance and
anxiety, positive and negative affectivity, self
‐esteem, and
sensation seeking (Luo, 2017).
Secondly, the similarity/attractiveness hypothesis states that
people are generally attracted to those who are similar to
themselves (Wee & Lee, 2017). Individuals assess their
characteristics and then select others who are similar. This
process works to reduce the potential degree of conflict in their
relationship (Byrne, 1971), or to reduce the psychological
discomfort that may arise from cognitive or emotional
differences (Lungeanu & Contractor, 2015). Studies that have
proven the similarity/attractiveness hypothesis include studies
into the following (among others) personality traits
(Bleda, 1974), attitudes (Yeong Tan & Singh, 1995), ethnic
backgrounds (Hu et al., 2008), voice features (Nass &
Brave, 2005), and also facial features (Bailenson et al., 2008)
with no conscious awareness of the assessors to the manipulation
of similarity.
Thirdly, perceiving familiar faces evokes positive feelings and,
on the contrary, unfamiliar ones provoke caution and fear (Cao,
Han, Hirshleifer, & Zhang 2011) or even hostility. An unfamiliar
object or organism (or a person with strange facial features) is a
potential source of danger for an organism striving for survival,
hence it becomes alert, cautious, and prepares for flight or fight
as needed. Fear of the unknown and xenophobia are considered
to be a fundamental fear (Carleton, 2016) and such a “setup”
provokes mostly negative emotions. Through this mechanism,
objects that are frequently encountered and do not represent a
danger evoke more positive reactions, are preferred and
considered to be more attractive than unfamiliar or unusual ones.
And finally, the theory of cognitive averaging states that subjects
organize and classify sensory information into categories (e.g.,
‟chairs”, “dogs”, or ‟faces” etc.). Cognitive averaging of the
individual examples within certain categories creates a central
representative of the category –
a ‟prototype”. An important
consequence of prototype formation is that subjects find the
prototype more attractive than any individual category member.
The reason for is that the prototype (due to its familiarity) is
easier (cognitively) to process (Pallet, Link, & Lee, 2010). Due
to this cognitive averaging mechanism
prototypes are often
preferred to individual exemplars of the categories of stimuli
(
Whitfield & Slatter, 1979; Martindane & Moore, 1988)
and for
example a face that is familiar (resembles one’s own face) is
perceived as more attractive than any individual face
(Rubenstein, Kalakanis, & Langlois, 1999).
A number of studies have proved the effect of the theories
mentioned that lead to mating with similar partners. Most of
them tested the similarity of existing pairs after partnerships of
various lengths. However, in many cases, the resemblance
observed in the personality of couples did not change with the
length of the partnership and – actually – some characteristics
seemed to correlate better in couples who had lived together for
a shorter period (Escorial & Martin-Buro, 2012). It has been
found that the attractiveness of similarity also extends to
non‐romantic friendships (McPherson, Smith‐Lovin, & Cook,
2001). It, therefore, seems that the similarity of personalities in
couples is not the outcome of a shared life, but is the result of an
active choice of partner carried out at the beginning of a
relationship.
Research concerning the assessment of human faces shows that
people are able to assess the characteristics of others from their
faces. Many authors (see e.g., Borkenau & Liebler, 1992, 1993;
Costa & McCrae, 1992; Goldberg, 1993; Penton-Voak, Pound,
Little, & Perrett, 2006) have proved the predictive value of facial
features for specific personality factors. More recently, most of
this research has employed the Big Five personality traits (for a
review, see Connolly, Kavanagh, & Viswesvaran, 2007) with the
main emphasis on conscientiousness and extraversion. This body
of evidence has also led to the creation of a variety of facial
composites that represent specific personalities (e.g., Big Five
traits – Penton-Voak, Pound, Little, & Perrett, 2006) or other
(e.g. intelligence – Kleisner, Chvátalová, & Flegr, 2014)
features. Further research also shows that the assessed degree of
- 41 -