What Does Your Selfie Say About You?
Lin Qiu* Jiahui Lu Shanshan Yang
Division of Psychology
Nanyang Technological University
Weina Qu Tingshao Zhu
Institute of Psychology
Chinese Academy of Sciences
Correspondent author*:
Lin Qiu, Division of Psychology, School of Humanities and Social Sciences, Nanyang
Technological University, 14 Nanyang drive, Singapore, 637332
Phone: +65 6513-2250
Fax: +65 6795-5797
Email: linqiu@ntu.edu.sg
Please cite this article as: Qiu, L., Lu, J., Yang, S., Qu, W., & Zhu, T. (2015). What
does your selfie say about you?. Computers in Human Behavior, 52, 443-449.
Highlights:
Selfies contain cues indicating personality traits.
A coding scheme for selfies was developed.
We found cues related to self-report and observer judgment of personality.
Observers made consistent judgment of personality traits from selfies.
Observers accurately predicted openness from selfies.
*Highlights (for review)
What Does Your Selfie Say About You?
Abstract
Selfies refer to self-portraits taken by oneself using a digital camera or a smartphone.
They become increasingly popular in social media. However, little is known about
how selfies reflect their owners' personality traits and how people judge others'
personality from selfies. In this study, we examined the association between selfies
and personality by measuring participants' Big Five personality and coding their
selfies posted on a social networking site. We found specific cues in selfies related to
agreeableness, conscientiousness, neuroticism, and openness. We also examined zero-
acquaintance personality judgment and found that observers had moderate to strong
agreement in their ratings of Big Five personality based on selfies. However, they
could only accurately predict selfie owners’ degree of openness. This study is the first
to reveal personality-related cues in selfies and provide a picture-coding scheme that
can be used to analyze selfies. We discussed the difference between personality
expression in selfies and other types of photos, and its possible relationship with
impression management of social media users.
Keywords
selfie, personality, zero-acquaintance judgment, photo, social media
*Manuscript without Author Details
Click here to view linked References
1. Introduction
"Selfie" was named the word of the year in 2013 by the Oxford English
Dictionary. It refers to a self-portrait picture taken by oneself using a digital camera or
a smartphone for posting on social networking sites. When taking a selfie, individuals
can view how they look like in the picture and decide what they want to show in the
picture. Millions of selfies have been posted on various social networking sites
(Unmetric, 2014). They have become a new medium for self-expression and self-
representation. While studies on social media have examined how personality is
related to the use of Facebook (Amichai-Hamburger & Vinitzky, 2010; Bachrach,
Kosinski, Graepel, Kohli, & Stillwell, 2012; Moore & McElroy, 2012; Gosling,
Augustine, Vazire, Holtzman, & Gaddis, 2011; Orchard, Fullwood, Galbraith, &
Morris, 2014; Ross et al., 2009; Seidman, 2013; Wilson, Gosling, & Graham, 2012)
and Twitter (Hughesa, Rowe, Batey, & Lee, 2012; Qiu, Lin, Ramsay, & Yang, 2012),
little is known about the relationship between personality and selfie. Do selfies reflect
their owners' personality traits? Can people predict others' personality based on their
selfies? Answers to these questions can improve our understanding of personality
expression and judgment in social media.
Past research has shown that traces of people’s personality can be found in their
environments and belongings. For example, extraverts’ offices are warm, decorated,
and inviting, conscientious individuals have neat and well-organized bedrooms, and
those who are open to experiences have a great variety of books and magazines in
their bedrooms (Gosling, Ko, Mannarelli, & Morris, 2002). Conscientious individuals
are less likely to wear high-top shoes, while emotionally stable individuals are more
likely to wear shoes with brand names (Gillath, Bahns, Ge, & Crandall, 2012).
Studies have also found cues such as facial expression and body posture in photos that
are related to personality (e.g., Borkenau, Brecke, Möttig, & Paelecke, 2009;
Naumann, Vazire, Rentfrow, & Gosling, 2009). However, these studies often use
portraits taken by others, but not participants themselves for the purpose of self-
expression. Research has shown that personality expression differs in different
contexts (Gosling et al., 2002). Therefore, personality expression in selfies is likely to
be different than those in other types of photos. Furthermore, selfies contain unique
cues that are not available in other types of photos. For example, duckface, a facial
expression made by pushing lips outward and upward to give the appearance of large
and pouty lips, is often seen in selfies but not other types of portraits. Such cues may
reveal new personality expression in photos. Therefore, we aim to identify
personality-related cues in selfies and examine how people express and judge
personality based on selfies.
2. Background research
2.1 Personality expression in photos
Past studies have shown that photos contain valid personality-related cues.
Nestler, Egloff, Küfner, and Back (2012) focused on standardized photographs and
found that extraversion is associated with attractiveness of face, openness is
associated with volume of lips, and conscientiousness is associated with femininity of
face. These cues are mainly about facial features, and cannot be changed by the user
when taking the pictures. Other studies used spontaneous photographs taken by
experimenters and found that extraversion was associated with cheerfulness and
smiling (Borkenau et al., 2009; Naumann et al., 2009) while narcissism was
associated with attractiveness, flashy clothing, and make up (Vazire, Naumann,
Rentfrow, & Gosling, 2008). In addition, extraverts stood in more energetic ways
while introverts stood in a tenser manner in full-body photos (Naumann et al., 2009).
While photos used in these studies contain cues that can be manipulated by the
participants (e.g., facial expression and body posture), they were not taken in a
naturalistic setting for the purpose of self-expression.
Several recent studies examined profile pictures in social media. Hall and
Pennington (2013) revealed that number of friends in Facebook profile picture was
associated with extraversion, and friendliness was associated with conscientiousness.
Ong et al. (2011) showed that self-rated attractiveness of profile pictures predicted
extraversion and narcissism. Krämer and Winter (2008) studied profile pictures on a
German social networking site, and found that extraverts tend to use photos with a
non-realistic style (e.g., altered color or graphically edited). The above studies
provided evidences of personality expression in photos in social media. However,
they did not focus on selfies.
Compared to other types of photos, selfies give individuals more freedom of
controlling their face visibility, emotional expression, and camera position. Therefore,
they may contain new cues, such as duckface and camera height, that are not available
in standardized photos. Furthermore, selfies are often posted on social media
platforms used for self-presentation (Mehdizadeh, 2010; Papacharissi, 2011). As the
motivation for self-expression and freedom of control have been found to result in
stronger cues for personality (Gosling et al., 2002), selfies may provide a better view
of their owners’ personality traits than other photos. However, studies have shown
that individuals are likely to be concerned about their online self-image and
manipulate their self-presentation to create socially desirable self-image (Ellison,
Heino, & Gibbs, 2006; Bazarova, Taft, Choi, & Cosley, 2013; Lin, Tov, & Qiu, 2014;
Qiu, Lin, Leung, & Tov, 2012; Strano, 2008). They have been found to promote
themselves and obtain positive feedback from their social networks via profile
pictures (Manago, Graham, Greenfield, & Salimkhan, 2008; Mehdizadeh, 2010;
Siibak, 2009; Zhao, Grasmuck, & Martin, 2008). This suggests that individuals may
create selfies that do not reflect their actual personality. Therefore, it is important to
examine which cues in selfies remain predictive of selfie owners’ true personality.
2.2 Zero-acquaintance personality judgment
An accumulating body of research indicates that personality can be judged by
unfamiliar others with reasonable accuracy. Such zero-acquaintance judgments
(Kenny & West, 2008) are made possible by the presence of personality-related cues,
such as facial expressions (Kenny, Horner, Kashy, & Chu, 1992), physical appearance
(Borkenau & Liebler, 1992; Naumann et al., 2009), choices of footwear (Gillath et al.,
2012), living environment (Gosling et al., 2002), musical preferences (Rentfrow &
Gosling, 2006), and linguistic patterns (Holleran & Mehl, 2008; Mehl, Gosling, &
Pennebaker, 2006; Qiu, Lin, Ramsay, & Yang, 2012).
Studies have shown that people can accurately judge personality traits based on
photos (Berry & Finch-Wero, 1993; Shevlin, Walker, Davies, Banyard, & Lewis,
2003). Findings demonstrated accurate judgment of extraversion from composite
facial images of extraverts or introverts (Little & Perrett, 2006), prediction of
agreeableness, extraversion, and neuroticism based on face images (Penton-Voak,
Pound, Little, & Perrett, 2006), and accurate judgments of trustworthiness,
competence, and aggressiveness after a 100-ms exposure to face portraits (Willis &
Todorov, 2006). An even shorter 50-ms exposure to a face was found to be enough
for accurate judgment of extraversion (Borkenau et al, 2009). Besides facial images,
Naumann et al. (2009) used full-body photographs as stimuli and found that
personality traits such as extraversion could be predicted.
A number of cues have been found to predict accurate judgment. For example,
attractiveness of face was used to predict IQ scores from black-and-white photos
(Zebrowitz, Hall, Murphy, & Rhodes, 2002). Clothing style was associated with the
prediction of openness, and smiling was related to the judgment of extraversion and
agreeableness in full-body photographs (Naumann et al., 2009). Fashionable and
stylish clothes, neat appearance, and attractiveness were the cues for accurate
narcissism judgment (Vazire et al., 2008). While the above studies identified cues
used in personality judgment, they were designed to examine the role of facial
expression and physical appearance. Their photographs were taken in standard
experimental settings and did not include contextual cues such as location information
and being alone. Since contextual cues can influence perception of personality and
emotion (Ito, Masuda, & Hioki, 2012; Ito, Masuda, & Li, 2013), it is important to
investigate how these cues in selfies are related to personality judgment.
2.3 The lens model of personality judgment
Brunswik’s (1956) lens model provides a useful framework for
conceptualizing and studying interpersonal judgment. It has been widely applied in
personality judgment research (e.g., Küfner, Back, Nestler, & Egloff, 2010; Nestler et
al., 2010; Rodriguez, Holleran, & Mehl, 2010). According to the model, a given
criterion variable (e.g., a personality trait such as extraversion) can be thought of as a
function of several observable cues (e.g., tendency to smile, physical attractiveness).
Meanwhile, the subjective judgment of that criterion variable (e.g., observer ratings of
extraversion) can also be considered as a function of the same cues. Cue validity is
the degree of association between a given cue and the criterion variable, with a
stronger correlation indicating higher validity. Cue utilization is the degree of
association between a given cue and the resulting judgment, with a stronger
correlation indicating greater utilization of that cue when forming personality
judgments. The lens model is particularly useful because it decomposes the notion of
accuracy – how closely the judgment matches the criterion variable – into two distinct
components: cue validity and cue utilization. For a personality judgment to be
accurate, a cue must be (a) related to the criterion variable, and (b) successfully
utilized. In essence, cues can be considered as mediators of the criterion-judgment
relationship.
The lens model offers an ideal platform for studying the relationships between
selfies, personality, and interpersonal perception. We adopted this model to examine
how personality is expressed in selfies and what cues people might use when making
personality judgments.
2.4 The present research
The goals of the present study are threefold. We aim to (1) examine if zero-
acquaintance personality judgments can be accurately made from selfies, (2) identify
valid cues in selfies associated with self-report personality traits, and (3) identify
potential cues observers may rely on to make personality judgments.
3. Method
3.1 Participants
Participants were recruited via two ways. We developed a software program and
crawled 1,953,485 users from Sina Weibo (a popular microbloging website similar to
Twitter in China). We then randomly selected 50,000 users and sent each user a
participation request. A total of 505 users participated in return for payment of
RMB30 (US$4.8) per person. The low response rate was likely due to the huge
amount of spam on Sina Weibo that made users frequently ignore participation
requests. Meanwhile, we recruited 107 Chinese students who were Sina Weibo users
from a large university in Singapore. Each student received S$5 (US$4.03) for their
participation.
3.2 Procedure
All participants completed a two-part online survey. The first part comprised of
the 44-item Big Five Personality Inventory (BFI; John, Donahue, & Kentle, 1991).
The second part asked participants about their Sina Weibo user names, usage
frequency, and demographic information (i.e., gender, age, country of residence, and
ethnicity).
Then, we downloaded the profile pictures of all participants. Two independent
raters identified which of these profile pictures were selfies and non-selfie portraits. A
total of 123 pictures were identified as selfies by both raters. Among their owners, 89
(72.4%) are females and 34 are males. Eleven (8.9%) are below 18 years old, 59
(48%) are between 18 and 20 years old, 44 (35.8%) are between 21 and 25 years old,
and 9 (7.3%) are above 26 years old. A total of 110 pictures were identified as non-
selfie portraits. Among their owners, 59 (53.6%) are females and 51 are males. Seven
(6.3%) are below 18 years old, 26 (23.6%) are between 18 and 20 years old, 52
(47.2%) are between 21 and 25 years old, and 25 (22.7%) are above 26 years old.
Overall, selfie owners were more likely to be females and younger than non-selfie
owners.
To code the selfies, we first selected picture-coding cues that are appropriate for
coding selfies from past research (Hall & Pennington, 2013; Krämer & Winter, 2008;
Nestler et al., 2012; Wang, 2012). Then, we added cues that are unique for selfies.
This results in a total of thirteen cues: duckface (0 = not duckface, 1 = duckface),
pressed lips (0 = not pressed lips, 1 = pressed lips), emotional positivity (0 = negative
emotion, 1 = neutral, 2 = positive emotion), eyes looking at the camera (0 = not
looking at camera, 1 = looking at the camera), camera height (0 = below head, 1 =
same level of head, 2 = above head), camera in front (0 = not in front, 1 = in front),
face visibility (0 = no face, 1 = part of face, 2 = complete face), amount of body (0 =
face only, 1 = include body from breast or shoulder up, 2 = include body from waist
up), alone (0 = not alone, 1 = alone), location information (0 = no location
information, 1 = have location information), public location (e.g., wilderness, city,
party, business setting; 0 = not public location, 1 = public location), private location
(e.g., bedroom, apartment; 0 = not private location, 1 = private location), and
photoshop editing (0 = no Photoshop editing, 1 = Photoshop editing). Two
independent raters coded these cues in the selfies. The averaged percentage agreement
of their coding was 90.81%. If an item received inconsistent coding from the two
raters, another rater re-coded the item and made the final judgment. Following Nestler
et al. (2012), good-looking of face was coded on a 5-point scale (1 = not at all, 5 =
very much) by the two raters to control for the effect of attractiveness of the selfie
owner. Its inter-rater reliability was .57, p < .001. The pair of coders’ ratings was
aggregated to form a composite rating for good-looking face.
Finally, eight undergraduate research assistants (1 male, 7 females; age: M = 21,
SD = 1.77) viewed each selfie and rated their impression of the selfie owner’s
personality using the same BFI that was used by the participants.
4. Results
4.1 Consensus and accuracy
Table 1 shows the Big Five personality traits of selfie owners. We found no
significant difference between selfie and non-selfie owners' personality traits after
controlling for age and gender (ps > .24). Intra-class correlations of single and
average observers were calculated to measure judgment consensus of selfie owners’
personality traits (Vazire & Mehl, 2008). Averaged observers’ ratings reached a
moderate
to strong consensus on all five personality dimensions. This suggested that observers
were consistent in predicting participants’ personality, and might utilize similar cues
for judgment. Extraversion showed the highest consensus, consistent with previous
zero-acquaintance judgment results based on pictures (Borkenau et al., 2009; Kenny,
Albright, Malloy, & Kashy, 1994; Kenny et al., 1992).
Regarding judgment accuracy, results showed significant correlation between
self-report and aggregated observers’ ratings on openness. This suggests that
Table 1 Self and observer rating of personality: consensus, accuracy, and vector correlation
Self-rating
ICC
accuracy
vector
correlation
M
SD
Cronbach's α
average
single
aggregate
single
Extraversion
3.27
.59
.69
.84***
.40***
.02
.01
.38
Agreeableness
3.59
.51
.63
.67***
.21***
.06
.03
.39
Conscientiousness
3.16
.52
.68
.69***
.22***
.10
.05
-.12
Neuroticism
3.10
.61
.67
.72***
.25***
.07
.04
.58*
Openness
3.64
.54
.71
.60***
.16***
.21*
.12
.70***
Note. Aggregated observer accuracy is the correlation between the aggregated observers’ rating and self-report personality.
Single observer accuracy is the average of correlations between a single observer’s rating and self-report personality. Vector
correlation is the correlations between cue-utilization correlations and cue-validity correlations after Fisher’s r-to-Z
transformation.
* p < .05, **p < .01, ***p < .001.
observers can accurately predict openness based on selfies, and this result is also
consistent with previous studies on judgment based on spontaneous pictures
(Naumann et al., 2009). However, our results showed that observers were not accurate
in predicting other four personality dimensions. In addition, when accuracy was
examined for the single observer, none of the five dimensions could be judged
accurately. This suggests that multiple observers are needed to improve judgment
accuracy.
4.2 Cue validity
We assessed cue validity by correlating participants’ self-report personality with
cues in the selfies (see Table 2). Extraversion was not related to any cue, different
from previous finding where extraversion was related to positive emotional
expression (Borkenau et al., 2009; Naumann et al., 2009). This is likely because
individuals tend to show positive emotion in their selfies due to their impression
management concerns, regardless of their degree of extraversion. Agreeableness was
Table 2 A Brunswik (1956) lens model analysis of judgments based on selfies: Cue-validity correlation
Cue-validity correlation
Extra.
Agree.
Cons.
Neur.
Open.
Gender
Age
duckface
-.07
-.12
-.03
.21*
-.16
.23**
-.22*
pressed lips
.02
.09
.13
-.09
-.06
-.17
.01
emotional positivity
.10
.18*
.00
-.06
.22*
.26**
.21*
eyes looking at the camera
-.14
-.03
.02
.06
-.08
.17
.16
camera height
-.05
-.20*
-.18
-.03
.04
.24**
.08
camera in front
.08
-.08
-.09
.15
.02
.15
-.13
face visibility
.00
-.02
.01
.00
-.03
-.03
.18*
amount of body
-.01
.02
-.08
.04
-.01
-.23*
.00
alone
-.08
-.03
-.15
.09
-.06
.19*
-.06
location information
.00
.13
-.14
-.13
-.02
-.26**
.07
public location
-.04
.13
.05
-.14
-.02
-.09
.07
private location
.03
.04
-.20*
-.04
-.01
-.22*
.03
Photoshop editing
-.07
.01
.06
.12
.10
.21*
-.13
Note. Cue-validity correlations indicate the correlations between cues and self-report personality. Extra., Extraversion;
Agree., Agreeableness; Cons., Conscientiousness; Neur., Neuroticism; Open., Openness. Gender: 0 = male, 1 = female.
* p < .05, **p < .01, ***p < .001.
associated with emotional positivity (.18), replicating past findings (Naumann et al.,
2009). It was also negatively associated with camera height (-.20), suggesting that
more agreeable individuals are more likely to take pictures from below.
Conscientiousness was negatively correlated to private location (-.20), suggesting that
more conscientious individuals are less likely to reveal their personal space in the
background. The avoidance of showing personal spaces reflects conscientious
individuals’ characteristics of being cautious (Costa & McCrae, 1992, 1996) and
concerned about their privacy (Junglas, Johnson, & Spitzmuller, 2008). Neuroticism
was related to duckface (.21), suggesting that neurotic individuals tend to make
duckface in their selfies. Openness was related to emotional positivity (.22), a
relationship that has not been documented in previous studies.
4.3 Cue utilization
We correlated aggregated observers’ ratings and selfie cues to identify possible
cues that observers used when judging personality (see Table 3). Past research has
Table 3 A Brunswik (1956) lens model analysis of judgments based on selfies: Cue-utilization correlation
Cue-utilization correlation
Extra.
Agree.
Cons.
Neur.
Open.
duckface
-.03
-.11
-.31**
.25**
-.14
pressed lips
-.19*
-.06
-.06
.16
-.22*
emotional positivity
.29**
.50**
.25**
-.40**
.21*
eyes looking at the camera
-.05
.24**
.06
.01
-.01
camera height
.13
.04
-.06
.00
.10
camera in front
.03
.10
-.03
.06
.06
face visibility
.14
.14
.08
-.21*
-.26**
amount of body
.14
-.03
.14
-.22*
-.03
alone
-.17
.03
-.12
.22*
.06
location information
-.07
.09
.30**
-.19*
-.06
public location
.06
-.01
.25**
-.13
.00
private location
-.12
.11
.14
-.11
-.07
Photoshop editing
-.01
-.03
-.20*
.09
.12
Note. Cue-utilization correlations indicate the correlations between cues and observer ratings. Extra.,
Extraversion; Agree., Agreeableness; Cons., Conscientiousness; Neur., Neuroticism; Open., Openness.
* p < .05, **p < .01, ***p < .001.
shown that extravert tend to be more sociable, talkative, and express more positive
emotion than introverts (House & Howell, 1992; McCrae & Costa, 1987; Qiu et al.,
2012). Ratings of extraversion were correlated with emotional positivity (.29). This is
consistent with the characteristics of extraversion, supporting the relationship between
smiling and judgment of extraversion in previous studies (Naumann et al., 2009).
Extraversion rating was also negatively related to the facial cue of pressed lips (-.19),
possibly because having pressed lips can be considered a sign of shyness.
Agreeable individuals are kind, cooperative and trusting, and they value social
affiliation (Bono & Judge, 2004; Nadkarni & Herrmann, 2010). Observers’ ratings of
agreeableness were associated with emotional positivity (.50), suggesting that
individuals showing more positive emotion in selfies were rated as more agreeable.
Agreeableness ratings were also associated with eyes looking at the camera (.24),
indicating that observers considered participants who had direct eye contact with them
as more agreeable than those who did not.
Conscientious individuals tend to be cautious, intolerant of ambiguity,
hardworking, and disciplined (Costa & McCrae, 1992, 1996). Ratings of
conscientiousness were associated with location information (.30) and public location
(.25), suggesting that observers rated those taking selfies in public as more
conscientious. Ratings of conscientiousness were also positively correlated with
emotional positivity (.25), and negatively correlated with Photoshop editing (-.20) and
duckface (-.31).
Neuroticism is associated with anxiety, moodiness, low self-esteem, and more
negative emotions (Judge, Bono, Ilies, & Gehardt, 2002; McCrae & Costa, 1987).
Therefore, it is reasonable for observers to negatively associate emotional positivity (-
.40) with their ratings of neuroticism. In addition, duckface (.25) and face visibility (-
.21) were related to the neuroticism ratings, suggesting that making duckface and not
showing full face were perceived as being moody. Observers also considered
participants who zoomed in on their faces as more neurotic. Therefore, their ratings of
neuroticism were negatively correlated with amount of body (-.22) and location
information (-.19), and positively correlated with being alone (.22).
Openness to experience is related to creativity, curiosity, risk-taking, and
preference for novelty and variety (Herrmann & Nadkarni, 2013; McCrae & Costa,
1987; Tetlock, 1983). Thus, it is reasonable that observers judged selfies with normal
full faces as lower degree of openness (-.26). Participants with pressed lips were rated
as less open (-.22), suggesting that making pressed lips was perceived as a facial
expression of closeness. Ratings of openness were also related to emotional positivity
(.21), supporting previous finding of smiling as a cue for openness judgment
(Naumann et al., 2009).
4.4 Sensitivity
The match between the pattern of cue utilization and cue validity indicates a
sensitivity of observers towards valid cues (Borkenau & Liebler, 1992; Funder &
Sneed, 1993). We performed vector correlations using the method proposed by
Funder and Sneed (1993). After Fisher’s r-to-z transformation, correlations of cue-
utilization and cue-validity were correlated across all the cues. Previous findings
suggest that traits that are accurately judged are associated with high vector
correlations (e.g., Back et al. 2010; Qiu et al, 2012). Our results reflected similar
patterns (see Table 1). Strong vector correlation was found for openness, indicating
that observers used valid cues to generate accurate judgment of openness. A moderate
vector correlation was found for neuroticism. The other three dimensions (i.e.,
extraversion, agreeableness, and conscientiousness) did not obtain significant vector
correlations.
We tested the mediating role of emotional positivity in self-other agreement for
openness, since it is the only cue correlated with both observer and self-rating.
Preacher and Hayes’ (2008) INDIRECT macro was used to test the mediation model.
The total effect of self-report openness on observers’ judgment was significant (B
=.10, SE =.04, p =.02). The indirect effect was significant (CI = [.0001, .0582],
excluding zero), while the direct effect of self-report openness on observers’
judgment was not significant (B =.08, SE =.04, p =.07). Overall, the mediation model
was significant, F (2, 120) = 4.62, p = .01, R2 = .07. Our results indicated that
emotional positivity fully mediated the accuracy for the judgment of openness.
4.5 Gender and age effect
Individuals with different gender and age varied in how they present themselves
in selfies (see Table 2). It is possible that the observed cue-validity correlations might
be contingent on these two variables. Thus, we calculated partial cue-validity
correlations by controlling gender and age. All initial significant cue-validity
correlations remained, suggesting that these cues were directly related to personality
traits.
Previous research suggested that observers might judge personality based on
stereotypes of age and gender (Gosling et al., 2002; Kenny et al., 1992). Thus, partial
correlations controlling for age and gender were calculated for cue-utilization to test
this possibility. Among the initial 18 significant correlations, three of them became
insignificant (i.e., photoshop editing and conscientiousness, face visibility and
neuroticism, and location information with neuroticism). This suggests that observers’
judgment were mainly based on photo cues rather than the stereotypes of gender and
age.
4.6 Effect of good-looking face
If observers’ judgment did not rely on stereotypes of gender and age, would
they rely on other properties of the participants, such as good-looking faces? To test
this possibility, we calculated partial cue-utilization by controlling the rating for
good-looking face. Results showed that two of the initial eighteen significant
correlations (i.e., the correlation between face visibility and neuroticism and the
correlation between face visibility and openness) became insignificant. This is likely
because face visibility is related to the perception of good-looking face. Overall, these
results indicated that observers judged personality mainly based on cues that were not
related to the attractiveness of the participants.
5. Discussion
The current study contributes to existing research on personality and social
media use by examining personality expression and judgment in the context of selfies,
a new form of self-portraits in social media. We applied the lens model (Brunswik,
1956) and identified cues that reflected selfie owners’ personality traits, and cues
associated with observers’ judgment of personality. This is the first study that
examines the relation between personality and selfies. It has important implications.
Our study shows that selfies reflect their owners’ personality traits. We
identified a number of personality-related cues. For example, emotional positivity
predicts agreeableness and openness, duckface indicates neuroticism, and private
location in the background indicates less conscientiousness. These cues reflect the
characteristics of their corresponding personality traits. While past study has shown
the connection between personality and photo-related activities such as number of
photo uploads and albums on Facebook (Eftekhar, Fullwood, & Morris, 2014), our
study revealed specific cues in self-portraits related to personality.
Although selfies contain cues that predict personality, observers could only
accurately judge selfie owners’ degree of openness. This is different from previous
findings where observers could form accurate prediction of extraversion,
agreeableness, openness, and neuroticism (marginally) from spontaneous full-body
photo (Naumann et al., 2009), and accurately judge extraversion, agreeableness, and
neuroticism from facial images without expression (Penton-Voak et al., 2006). The
poor judgment found in our study was reflected in the asymmetry of cue-validity and
cue-utilization, suggesting that observers used invalid cues to judge personality. We
found that emotional positivity was used to judge all five personality traits, a
phenomenon that has been documented in past research (Naumann et al., 2009).
Several new cues were found to be related to personality judgment. For example,
duckface was related to the judgment of conscientiousness and neuroticism, pressed
lips were associated with the judgment of extraversion and openness, public location
in the background was related to the judgment of conscientiousness, and being alone
was related to neuroticism ratings.
Why do selfies contain limited personality-related cues and are difficult for
accurate zero-acquaintance personality judgment? There could be several reasons.
First, selfies allow individuals to have full control of their appearance. Individuals can
easily manipulate their facial expression and eye contact to appear different from how
they normally look. Second, selfies are often taken for sharing on social networking
sites. Previous research on impression management found that individuals could
accurately perceive norms, expectations, and social desirability from their social
networks (Siibak, 2009). It is possible that selfies are manipulated to present a
positive social image, so that typical associations such as the correlation between
smiling and extraversion (Naumann et al., 2009) become invalid. Thirdly, since
selfies are taken by individuals themselves, most of them only contain faces. This
prevents important cues, such as body posture and style of clothing, to appear in the
picture. Previous studies showed that standing pose provides information about
extraversion (Naumann et al., 2009) and clothing styles are indicative of extraversion
and conscientiousness (Albright, Kenny, & Thomas, 1988; Kenny et al., 1992;
Naumann et al., 2009). Thus, the lack of informative cues in selfies might result in
poor accuracy. However, it is important to note that even with possible impression
management strategies, people still inadvertently leave cues that predict their
agreeableness, conscientiousness, neuroticism, and openness.
Our research has important practical implications. With selfies becoming
extremely popular, there is great interest in understanding how they reflect personality.
By identifying valid cues related to selfie owners' personality traits, our research
provides important information for future work to improve the accuracy of human or
machine prediction of personality from selfies. For example, computer programs can
be developed to detect duckface to help predict neuroticism. One limitation of the
current study is that we used photos from a microblogging website. As different
online social networking sites have different user characteristics and usage patterns
(Wilson, Gosling, & Graham, 2012), future research needs to examine if our findings
can be generalized to other social networking sites.
6. Conclusion
The current study extends research on personality expression and judgment by
examining selfies, a new form of self-portraits in social media. We identified cues in
selfies that are related to selfie owners' degree of agreeableness, conscientiousness,
neuroticism, and openness. These cues included facial cues such as duckface and
emotion, and contexual cues such as background location. In addition, we examined
zero-acquaintance judgement and found that observers had moderate to strong
agreement on their prediction of all Big Five personality traits based on selfies.
However, they could only accurately predict selfie owners’ degree of openness. Our
study is the first to reveal personality-related cues in selfies, and suggests that the
difference between personality expression in selfies and other types of photos might
be due to impression management of social media users. We provided the first coding
scheme specific for selfies. Future studies in psychology, communication, and
human–computer interaction can use it to process selfies and further understand how
they reflect users' characteristics and psychological processes.
Reference
Albright, L., Kenny, D. A., & Thomas, E. M. (1988). Consensus in personality
judgments at zero acquaintance. Journal of Personality and Social Psychology,
55(3), 387–395.
Amichai-Hamburger, Y., & Vinitzky, G. (2010). Social network use and personality.
Computers in Human Behavior, 26(6), 1289–1295.
Bachrach, Y., Kosinski, M., Graepel, T., Kohli, P., & Stillwell, D. (2012). Personality
and patterns of Facebook usage. In Proceedings of the 3rd annual ACM Web
science conference on – WebSci ‘12 (pp. 24–32). New York, USA: ACM Press.
Back, M. D., Stopfer, J. M., Vazire, S., Gaddis, S., Schmukle, S. C., Egloff, B., et al.
(2010). Facebook profiles reflect actual personality, not self-idealization.
Psychological Science, 21(3), 372–374.
Bazarova, N. N., Taft, J. G., Choi, Y. H., & Cosley, D. (2013). Managing impressions
and relationships on Facebook: Self-presentational and relational concerns
revealed through the analysis of language style. Journal of Language and Social
Psychology, 32(2), 121-141.
Berry, D. S., Finch-Wero, J. L.(1993). Accuracy in face perception: A view from
ecological psychology. Journal of personality, 61(4), 497-520.
Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional
leadership: a meta-analysis. Journal of applied psychology, 89(5), 901-910.
Borkenau, P., Brecke, S., Möttig, C., & Paelecke, M. (2009). Extraversion is
accurately perceived after a 50-ms exposure to a face. Journal of Research in
Personality, 43(4), 703-706.
Borkenau, P., & Liebler, A. (1992). Trait inferences: Sources of validity at zero
acquaintance. Journal of Personality and Social Psychology, 62(4), 645-657.
Brunswik, E. (1956). Perception and the representative design of psychological
experiments. Berkeley: University of California Press.
Costa, P. T., & McCrae, R. R. (1992). Neo personality inventory–revised (neo-pi-r)
and neo five-factor inventory (neo-ffi) professional manual. Odessa, FL:
Psychological Assessment Resources.
Costa, P. T., & McCrae, R. R. (1996). Toward a new generation of personality
theories: Theoretical contexts for the five-factor model. The five factor model of
personality: Theoretical perspectives. Hrsg.: JS Wiggins. New York, 51-87.
Ellison, N. B., Heino, R. D., & Gibbs, J. L. (2006). Self-presentation processes in the
online dating environment. Journal of Computer-Mediated Communication, 11(2),
415–441.
Funder, D. C., & Sneed, C. D. (1993). Behavioral manifestations of personality: An
ecological approach to judgmental accuracy. Journal of Personality and Social
Psychology, 64(3), 479–490.
Gillath, O., Bahns, A. J., Ge, F., & Crandall, C. S. (2012). Shoes as a source of first
impressions. Journal of Research in Personality, 46(4), 423-430.
Gosling, S. D., Augustine, A. A., Vazire, S., Holtzman, N., & Gaddis, S. (2011).
Manifestations of personality in online social networks: Self-reported Facebook-
related behaviors and observable profile information. Cyberpsychology, Behavior,
and Social Networking, 14(9), 483–488.
Gosling, S. D., Ko, S. J., Mannarelli, T., & Morris, M. E. (2002). A room with a cue:
personality judgments based on offices and bedrooms. Journal of personality and
social psychology, 82(3), 379-398.
Hall, J. A., & Pennington, N. (2013). Self-monitoring, honesty, and cue use on
Facebook: The relationship with user extraversion and conscientiousness.
Computers in Human Behavior, 29(4), 1556-1564.
Herrmann, P., & Nadkarni, S. (2013). Managing strategic change: the duality of CEO
personality. Strategic Management Journal, 35(9), 1318-1342.
Holleran, S. E., & Mehl, M. R. (2008). Let me read your mind: Personality judgments
based on a person’s natural stream of thought. Journal of Research in Personality,
42(3), 747–754.
House, R. J., & Howell, J. M. (1992). Personality and charismatic leadership. The
Leadership Quarterly, 3(2), 81-108.
Hughesa, D. J., Rowe , M., Batey , M., & Lee, A. (2012). A tale of two sites: Twitter
vs. Facebook and the personality predictors of social media usage. Computers in
Human Behavior, 28(2), 561–569.
Ito, K., Masuda, T., & Hioki, K. (2012). Affective information in context and
judgment of facial expression: Cultural similarities and variations in context effects
between North Americans and East Asians. Journal of Cross-Cultural Psychology,
43(3), 429-445.
Ito, K., Masuda, T., & Li, L. M. W. (2013). Agency and facial emotion judgment in
context. Personality and Social Psychology Bulletin, 39(6), 763-776.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991).The Big Five inventory –
versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of
Personality and Social Research.
Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and
leadership: a qualitative and quantitative review. Journal of applied
psychology, 87(4), 765-780.
Junglas, I. A., Johnson, N. A., & Spitzmüller, C. (2008). Personality traits and
concern for privacy: an empirical study in the context of location-based
services. European Journal of Information Systems, 17(4), 387-402.
Kenny, D. A., Albright, L., Malloy, T. E., & Kashy, D. A. (1994). Consensus in
interpersonal perception: acquaintance and the big five. Psychological
bulletin, 116(2), 245-258.
Kenny, D. A., Horner, C., Kashy, D. A., & Chu, L. C. (1992). Consensus at zero
acquaintance: Replication, behavioral cues, and stability. Journal of personality
and social psychology, 62(1), 88-97.
Kenny, D. A., & West, T. V. (2008). Self-perception as interpersonal perception. In J.
V. Wood, A. Tesser, & J. G. Holmes (Eds.), The self and social relationships (pp.
119–137). New York: Psychology Press.
Krämer, N. C., & Winter, S. (2008). Impression management 2.0: The relationship of
self-esteem, extraversion, self-efficacy, and self-presentation within social
networking sites. Journal of Media Psychology: Theories, Methods, and
Applications, 20(3), 106-116.
Küfner, A. C. P., Back, M. D., Nestler, S., & Egloff, B. (2010). Tell me a story and I
will tell you who you are! Lens model analyses of personality and creative writing.
Journal of Research in Personality, 44(4), 427–435.
Lin, H., Tov, W., & Qiu, L. (2014). Emotional disclosure on social networking sites:
The role of network structure and psychological needs. Computers in Human
Behavior, 41, 342-350.
Little, A. C., & Perrett, D. I. (2006). Using composite images to assess accuracy in
personality attribution to faces. British Journal of Psychology, 98(1), 111-126.
Mehdizadeh, S. (2010). Self-Presentation 2.0: Narcissism and Self-Esteem on
Facebook. Cyberpsychology, Behavior, and Socla Networking, 13(4), 357-364.
Manago, A. M., Graham, M. B., Greenfield, P. M., & Salimkhan, G. (2008). Self-
presentation and gender on MySpace. Journal of Applied Developmental
Psychology, 29(6), 446-458.
McCrae, R. R., & Costa, P. T. (1987). Validation of the five-factor model of
personality across instruments and observers. Journal of personality and social
psychology, 52(1), 81-90.
Mehl, M. R., Gosling, S. D., & Pennebaker, J. W. (2006). Personality in its natural
habitat: Manifestations and implicit folk theories of personality in daily life.
Journal of Personality and Social Psychology, 90(5), 862–877.
Moore, K., & McElroy, J. C. (2012). The influence of personality on Facebook usage,
wall postings, and regret. Computers in Human Behavior, 28(1), 267–274.
Nadkarni, S., & Herrmann, P. O. L. (2010). CEO personality, strategic flexibility, and
firm performance: the case of the Indian business process outsourcing
industry. Academy of Management Journal, 53(5), 1050-1073.
Naumann, L. P., Vazire, S., Rentfrow, P. J., & Gosling, S. D. (2009). Personality
judgments based on physical appearance. Personality and Social Psychology
Bulletin, 35(12), 1661-1671.
Nestler, S., Egloff, B., Küfner, A. C., & Back, M. D. (2012). An integrative lens
model approach to bias and accuracy in human inferences: Hindsight effects and
knowledge updating in personality judgments. Journal of personality and social
psychology, 103(4), 689-717.
Ong, E. Y., Ang, R. P., Ho, J., Lim, J. C., Goh, D. H., Lee, C. S., & Chua, A. Y.
(2011). Narcissism, extraversion and adolescents’ self-presentation on
Facebook. Personality and Individual Differences, 50(2), 180-185.
Orchard, L. J., Fullwood, C., Galbraith, N., & Morris, N. (2014). Individual
differences as predictors of social networking. Journal of Computer Mediated
Communication, 19(3), 388–402.
Papacharissi, Z. (2011). A networked self: Identity, community, and culture on social
netowork. New York, NY: Routledge.
Penton-Voak, I. S., Pound, N., Little, A. C., & Perrett, D. I. (2006). Personality
judgments from natural and composite facial images: More evidence for a “kernel
of truth” in social perception. Social Cognition, 24(5), 607-640.
Qiu, L., Lin, H., Leung, A. K., & Tov, W. (2012). Putting their best foot forward:
Emotional disclosure on Facebook. Cyberpsychology, Behavior, and Social
Networking, 15(10), 569-572.
Qiu, L., Lin, H., Ramsay, J., & Yang, F. (2012). You are what you tweet: Personality
expression and perception on twitter. Journal of Research in Personality, 46(6),
710-718.
Rentfrow, P. J., & Gosling, S. D. (2006). Message in a ballad the role of music
preferences in interpersonal perception. Psychological Science, 17(3), 236–242.
Rodriguez, A. J., Holleran, S. E., & Mehl, M. R. (2010). Reading between the lines:
The lay assessment of subclinical depression from written self-descriptions.
Journal of Personality, 78(2), 575–598.
Ross, C., Orr, E. S., Sisic, M., Arseneault, J. M., Simmering, M. G., & Orr, R. R.
(2009). Personality and motivations associated with Facebook use. Computers in
Human Behavior, 25(2), 578–586.
Seidman, G. (2013). Self-presentation and belonging on Facebook: How personality
influences social media use and motivations. Personality and Individual
Differences, 54(3), 402–407.
Shevlin, M., Walker, S., Davies, M. N., Banyard, P., & Lewis, C. A. (2003). Can you
judge a book by its cover? Evidence of self–stranger agreement on personality at
zero acquaintance. Personality and Individual Differences,35(6), 1373-1383.
Siibak, A. (2009). Constructing the Self through the Photo selection – Visual
Impression Management on Social Networking Websites. Cyberpsychology:
Journal of Psychosocial Research on Cyberspace, 3(1), article 1.
Strano, M. M. (2008). User descriptions and interpretations of self-presentation
through Facebook profile images. Cyberpsychology: Journal of Psychosocial
Research on Cyberspace, 2, article 5.
Tetlock, P. E. (1983). Accountability and complexity of thought. Journal of
Personality and Social Psychology, 45(1), 74-83.
Unmetric (2014). Year of the selfie. Retrieved from
http://blog.unmetric.com/2014/02/year-of-the-selfie/.
Vazire, S., & Mehl, M. R. (2008). Knowing me, knowing you: The accuracy and
unique predictive validity of self-ratings and other-ratings of daily behavior.
Journal of Personality and Social Psychology, 95(5), 1202–1216.
Vazire, S., Naumann, L. P., Rentfrow, P. J., & Gosling, S. D. (2008). Portrait of a
narcissist: Manifestations of narcissism in physical appearance. Journal of
Research in Personality, 42(6), 1439-1447.
Wang, Y. H. (2012). Taiwanese girls’ self-portraiture on a social networking
site (Doctoral dissertation) Retrieved from
http://core.kmi.open.ac.uk/download/pdf/9994909.pdf.
Willis, J., & Todorov, A. (2006). First impressions making up your mind after a 100-
ms exposure to a face. Psychological science, 17(7), 592-598.
Wilson, R. E., Gosling, S. D., & Graham, L. T. (2012). A review of Facebook
research in the social sciences. Perspectives on Psychological Science, 7(3), 203-
220.
Zebrowitz, L. A., Hall, J. A., Murphy, N. A., & Rhodes, G. (2002). Looking smart
and looking good: Facial cues to intelligence and their origins. Personality and
Social Psychology Bulletin, 28(2), 238-249.
Zhao, S., Grasmuck, S., & Martin, J. (2008). Identity construction on Facebook:
Digital empowerment in anchored relationships. Computers in human
behavior, 24(5), 1816-1836.
Acknowledgements
This work was supported by Singapore Ministry of Education AcRF Tier 1 Grant
RGT 37/13 awarded to the first author.
Acknowledgements