Abstract
Given the widespread phenomenon of selfies, numerous studies are examining the motivations behind taking and sharing selfies. The current paper suggests an additional possible motivation, namely, decreasing death anxiety. People are motivated to decrease their death anxiety by preserving a fake feeling of immortality. One known way to achieve this goal is by using photography. Therefore, we suggest that selfie behaviors are a way to fulfill the need to remain immortal. A hundred undergraduate students (Mage = 22.33) answered self-reported questionnaires regarding selfie motivations, selfie-taking frequency, selfie-sharing frequency, and death anxiety. All of those selfie measurements were indeed positively related to death anxiety. Moreover, many previous studies suggested that narcissism motivates selfie behaviors. In an exploratory approach, we examined whether death anxiety mediates this relationship. Indeed, death anxiety fully mediated the relationships between narcissism and selfie motivations and between narcissism and selfie-taking frequency, suggesting that the well-documented association between selfie behaviors and narcissism might be driven by death anxiety. Those preliminary results indicate that death anxiety is associated with selfie behaviors, opening new avenues for understanding the motivations underlying selfie behaviors.
Introduction
A “selfie” is a self- photograph, usually taken using a cellphone or a webcam (Weiser, 2015). Nowadays, it has become such a common phenomenon, that it has even been identified as a “selfie epidemic” (Reilly et al., 2019), an epidemic that is being encouraged by the “selfie culture” (Maddox et al., 2020). Indeed, in 2019, about 650 billion selfies were taken per day (Reilly et al., 2019).
Several studies have attempted to examine what lies behind the need to take and share selfies. The vast majority of them suggested that selfie behaviors are related to narcissism, a trait that has been defined as a grandiose sense of self-importance, accompanied by feelings of entitlement (Brown et al., 2009). Narcissism was found to be related to selfie-taking (Ansari & Azhar, 2022; Kaur & Vig, 2016; Koterba et al., 2021; McCain et al., 2016; Singh & Tripathi, 2017), selfie posting to social network sites (Shane-Simpson et al., 2019; Singh et al., 2018; Sung, et al., 2016; Weiser, 2015), and selfie editing (Wang, 2019). Individuals higher on narcissism reported selfie-taking and selfie posting as a more favorable activity than non-narcissistic individuals (Halpern et al., 2016; Lee & Sung, 2016). They were also more emotionally aroused as to whether they received any feedback when a selfie was uploaded (Nash et al., 2019). From a different angle, when laypersons are being asked if their selfie behaviors are arousing narcissism, about 55% of replies indicate the lay assumption that selfie behaviors are connected to narcissistic behaviors (Wickel, 2015).
Other studies suggested different motivations for selfie behaviors. Several implied that selfie behaviors are simply another form of behavior that fulfills human needs (Frosh, 2015; Murray, 2015). Other indicated specific human needs, such as the need for self-approval, feelings of belonging, the need to document personal experiences (Etgar & Amichai-Hamburger, 2017), the need to represent oneself using a visual image of any kind (Frosh, 2015; Kumar, 2018), or the need for self-expression (Iqani & Schroeder, 2016).
The current paper would like to suggest another human need that selfies can possibly fulfill. This is the need to remain immortal, or to avoid death anxiety. Terror Management Theory (TMT) suggests that one of our strongest motivators is our apprehension concerning our death (Greenberg et al., 1986). To live and function with such a threat, we must defend ourselves from this certain end. We do so by believing that some aspects of ourselves will continue to live after our physical disappearance. Several types of behaviors or beliefs are considered an expression of this need to remain immortal, such as having children or believing in life after death (Burke et al., 2010). Indeed, numerous studies found that an increment in mortality salience is positively related to these defensive behaviors or beliefs (for a review see: Burke et al., 2010).
One of the defensive behaviors that are being used to achieve feelings of immortality is photography. Photos are used for “imparting a tenuous immortality […] by providing an identity, a context of belongingness, to one’s descendants” (Csikszentmihalyi & Halton, 1981, p. 69), and they are assumed to preserve individuals’ need for immortality (Batchen, 2006; Bathes, 1981). This assumption goes beyond Western cultures, as the Chinese also believe that there is a connection between photography and immortality (Li et al., 2019).
And nowadays, photography is literally on the tip of our fingers: Mobile photography is common and easy to use (Palmer, 2014). The ease of photo taking and sharing helps us preserve them and transform them into public or private memories (Shanks & Svabo, 2014). Interestingly, selfie taking and sharing are two of the major “digital immortality” practices, suggesting how to preserve one’s digital self after death (Turchin, 2018). Different studies have also supported this notion: people in critical situations or those who were facing an almost certain death, such as a sinking ferry or a mass shooting, took large numbers of selfies during those situations (Hampton, 2020; Hjorth & Cumiskey, 2018). Similarly, selfies have become extremely popular during visits to the Auschwitz death camp, a place that undoubtedly increases mortality awareness (Feldman & Musih, 2023).
The current study goes beyond those studies by examining the relationships between selfie behaviors and death anxiety as a general tendency. This will allow us to examine whether a similar association also occurs in less extreme situations. To do so, the current study uses more comprehensive measurements: Selfie behaviors are measured using general scales for selfie motivations, frequency of selfie-taking, and frequency of selfie-posting; Death anxiety is measured as a general tendency, using a measurement that examines it as a trait. Given the aforementioned idea, suggesting that selfies are an easy way to increase the feeling of preserved immortality, we will hypothesize that death anxiety is positively related to selfie behaviors.
Interestingly to the selfie research area, the need to feel immortal is also an essential characteristic of narcissistic behavior and personality (Grunberger, 1979). In Freud’s book On narcissism (Freud, 1955, original work published 1914), he suggested that the ego’s deepest fantasy to be immortal versus the unfortunate truth that it is impossible creates a major dissonance within the narcissist’s experience. Since then, many studies have found support for this notion, suggesting that aspiration to immortality is a well-known feature of narcissism (Akhtar & Thomson, 1982; Hirsch, 2014; Kohut, 1971). This aspiration might take many different forms, such as that narcissist individuals frequently attempt to ignore well-defined time schedules (Kernberg, 2008), have religious beliefs about life after death, believe in mysticism (Noll & Rosenbaum, 2014), or want to be parents in order to leave something behind (Newton et al., 2014).
Since narcissism is highly related to the need to preserve mortality as well as to selfie behaviors, we would like to add an exploratory hypothesis, suggesting that mortality anxiety may serve as a mediator of the relationship between selfie behaviors and narcissism. While there is a well-established relationship between selfie behaviors and narcissism, its underlying mechanism remains unclear. The motivation to increase the sense of immortality might be at least one of the missing pieces of this puzzle.
Method
Participants
One hundred and eleven psychology undergraduates from a large Israeli university were recruited through an internal Web site dedicated to recruiting students for on-campus experiments. All participants received course credit for their participation. Of them, we excluded participants who had not completed the questionnaire (n = 6). Additionally, at the beginning of the questionnaire, participants were asked a filter question: “the term “selfie” refers to self-photographing via a smartphone. Are you taking selfies?”. Participants who answered “no” to this question (n = 5) were also excluded from the sample. The final sample included one hundred participants. Of them, 84 identified as women (Mage = 22.33, SDage = 1.63).
Measurements
Narcissism
Narcissism was measured using the NPI-16 (Ames et al., 2006). This is a 16-item scale, where each item contains two contrasting sentences with a 5-point scale between them. For instance, one sentence says, “I like to be the center of attention,” whereas its opposing sentence says, “I prefer to blend in with the crowd”. For each item, participants were asked to choose their place on the scale between these two sentences. Answers to the 16 items were averaged to create a scale, in which a higher score indicated a higher narcissism level (Cronbach’s α = .821).
Death Anxiety
Death anxiety was measured by the Fear of Personal Death Scale (Florian & Kravetz, 1983). The scale contains 31 items, each describing a reason for fear of death. Participants indicate how far death frightens them because of the reason described in each item. For example, “I’m afraid of death because I’m afraid I’ll be forgotten”. Answers were given on a 7-point scale ranging from “totally incorrect for me” to “totally correct for me”. Answers were averaged to create a death anxiety variable, where a higher score indicates a higher level of death anxiety (Cronbach’s α = .945).
Selfie Motivations
Selfie motivations measurement, which includes both the motivations for selfie-taking and posting, was measured by 15 items, each of which asked about a different motivation to produce or upload a selfie. For example, “I’m uploading a selfie because I want to receive ‘likes’”. Participants were asked whether they agreed or disagreed with each motivation on a scale ranging from 1 (definitely disagree) to 5 (definitely agree). A higher score indicates a higher selfie motivation. The selfie motivations scale was calculated by averaging all fifteen items (Cronbach’s α = .885).
Selfie-taking frequency
Participants were asked how frequently they take selfies in ten different situations, such as “alone”, “with friends”, “during sports events”, or “at parties”. The scale to each answer runs from: 1- more than once a day, to 10- never. Participants could also choose “N/A”. Before analyzing the data, we turned N/A answers into missing data and reversed the scale so that a higher score would represent a greater selfie-taking frequency. The selfie-taking frequency scale was calculated by averaging across these ten items (Cronbach’s α = .883).
Selfie Sharing Frequency
To measure the frequency of selfie sharing, participants were asked how frequently they are sharing their selfies with different crowds, such as “Uploading to Facebook” or “sending to friends.” The scale to each answer ranges from: 1- more than once a day, to 10- never. Participants could also choose “N/A”. One item stated “not sending the selfie”, and it has not been calculated into this scale. Two additional items, “uploading to a social network site that hasn’t been mentioned” and “other:” were not included as part of the final scale as most of the responses chose the N/A option. Before analyzing the data, we turned N/A answers into missing data and reversed the scale so that a higher score would represent a higher selfie posting frequency (Cronbach’s α = .822).
Control variables
Rosenberg’s (1979) self-esteem scale, as well as the Mini-IPIP big-five Scales (Donnellan et al., 2006) were also measured during this study as possible controls; However, none of these six variables was related to any of the selfie measurements (all p’s > .134), therefore, we will not discuss them further.
Procedure
Participants arrived at the lab to participate in a “selfie motivations” study. They used the lab’s computers to open Qualtrics.com and fill out those questionnaires. They were first asked to complete all the questionnaires related to selfie behaviors. Then, they answered the self-esteem, big five, and narcissism questionnaires, while the death anxiety questionnaire always appeared at the end. Before answering it, each participant watched a short neutral movie, used as a filer task. Participants were then thanked and debriefed.
Results
Descriptive Statistics and Correlations
Means, Standard Deviations, and Correlation Matrix for all Variables.
*p < .05; **p < .001.
Selfie Behaviors and Death Anxiety
Following our main hypothesis, and as can be seen in Table 1, correlations were found between all the selfie behaviors and death anxiety; selfie motivations (r = .42, p < .001), selfie-taking (r = .359, p < .001), and selfie sharing (r = .239, p = .017) were all related to death anxiety.
Death Anxiety as a Mediator
Following our exploratory hypothesis, we tested whether death anxiety might mediate the relationship between selfie behaviors and narcissism. Mediation models were conducted using bootstrapping with 95% confidence intervals based on 5000 random samples, at PROCESS macro for SPSS (Hayes, 2013, model 4).
The first model examined whether death anxiety mediated the relationship between narcissism and selfie motivations. It has been found that death anxiety fully mediated this relationship: The indirect effect was significant (Beta = .12, SE = .058, 95%CI [0.014, 0.24]). At the same time, adding death anxiety to the model led to an insignificant direct effect between narcissism and selfie motivation (Beta = .15, SE = .12, 95%CI [−.09, .39]). Path coefficients can be found in Figure 1. Since the number of participants is lower than recommended for mediation analysis, we examined the post-prior power of the current analysis by using Schoemann et al. (2017) Monte Carlo Power Analysis for Indirect Effects calculation tool. The indirect effect power seems to be mediocre, yet sufficient (Monte Carlo Power = 0.63). The Mediation model of narcissism on selfie motivations, mediated by death anxiety. *p < .05; **p < .001.
The second model examined whether death anxiety mediated the relationship between narcissism and the frequency of selfie-taking. Similarly to the first model, it was found that death anxiety fully mediated this relationship: The indirect effect was significant (Beta = .21, SE = .12, 95%CI [0.02,0.49]). At the same time, adding death anxiety to the model led to an insignificant direct effect between narcissism and selfie-taking (Beta = .52, SE = .28, 95%CI [−.021,1.07]). Path coefficients can be found in Figure 2. Here as well, we conducted a Monte Carlo Power Analysis to examine the post-prior power of this indirect effect (Schoemann et al., 2017). The indirect effect has a power of 0.56, suggesting a medicore, yet sufficient power to the model. The Mediation model of narcissism on selfie-taking, mediated by death anxiety. *p < .05.
We aimed to also explore whether death anxiety mediated the relationship between narcissism to selfie sharing. However, interestingly, this relationship is insignificant (r = .146, p = .147). Therefore, we did not explore this mediation model.
Discussion
Death anxiety was correlated to selfie motivations, selfie-taking frequency, and selfie sharing frequency, suggesting that the more people were aware (and afraid) of their death, the more they were taking and sharing selfies, as well as motivated to engage in such behaviors. These results suggest that death anxiety should be considered as another motivation to engage in different selfie behaviors (e.g., selfie taking, selfie sharing). This idea aligns with TMT, suggesting that the motivation to remain immortal is a strong motivation, expressed by many different attempts to “leave something behind” (Burke et al., 2010). Photographs have always been an easy way to achieve this goal (Batchen, 2006; Bathes, 1981; Csikszentmihalyi & Halton, 1981), and our research yields initial evidence that selfies can also fulfill this need.
Death anxiety was also found to fully mediate the relationship between narcissism and selfie motivations, as well as the relationship between narcissism and selfie-taking. In other words, this suggests that the well-documented relationship between narcissism and selfie behaviors (e.g., Halpern et al., 2016; Shane-Simpson, 2019; Singh et al., 2018; Sung et al., 2016; Wang, 2019; Weiser, 2015) might be underlined by death anxiety. Precisely, individuals with higher levels of narcissism have a greater degree of death anxiety (e.g., Akhtar & Thomson, 1982; Hirsch, 2014; Kernberg, 2008), which in turn relates to a higher number of selfie behaviors. These results raise the possibility that it is not the narcissism by itself that relates to selfie behaviors, but the death anxiety that accompanies the narcissistic experience. This might explain why narcissism was not found to be related to selfie behaviors when other selfie motivations were explored (Etgar & Amichai-Hamburger, 2017).
To the best of our knowledge, the current preliminary study uncovers an effect that hasn’t been discussed elsewhere. On the other hand, such preliminary findings always come with a price, and many questions regarding this effect remain open. For example, behind the general measurements of “selfie motivations”, “frequency of selfie-taking”, and “frequency of selfie-posting”, various sorts of behaviors or motivations were measured. While those measurements were inclusive enough to examine general selfie behaviors, further studies are needed to understand whether different selfie behaviors are linked differently to death anxiety. For instance, it makes sense to assume that those who are higher in death anxiety will take less “risky” selfies (i.e., selfies that are taken from risky places such as cliff edges or water bodies, see for example, Weiler et al., 2021), but will take more selfies in day-to-day situations, that individuals who are lower in death anxiety may perceive as less important to preserve.
The effect reported here was based on dispositional traits and measurements. Notably, it is similar to the effect that was found in real-life close-to-death situations (Hampton, 2020; Hjorth & Cumiskey, 2018). However, life-threatening situations are different in magnitude from “normal” day-to-day situations. It will therefore be interesting to examine this effect from a different angle by replicating it in a controlled experiment using mortality salience manipulations (for a review, see Burke et al., 2010).
This study has several limitations that must be discussed. First, this study suffers from a small sample size, which might affect its statistical power. Therefore, we suggest perceiving this research effort as an initial stage that requires further investigation. Second, while using undergraduates as participants is a common practice, this population’s representation might be somewhat in doubt (Henrich et al., 2010). This is especially important for technology-related issues, that could be age-dependent (Prensky, 2001). Future studies should examine the association between death anxiety and selfie behaviors across different populations. Third, this study is correlational; therefore, our ability to conclude causality is limited. Further studies should use experimental or longitudinal research methods to establish causality. This point is specifically crucial when it comes to the mediation models demonstrated in this work. While they have been suggested as an exploratory approach, future studies should consider the causal chains involved in such models by using a longitudinal exploration and a sufficient number of participants.
In conclusion, this study is the first to reveal a new motivation for selfie behaviors: death anxiety. Moreover, it supplies initial indications that death anxiety is also a mediator in the relationships between narcissism and selfie behaviors. As we are currently living in the age of selfies, revealing another motivation underlying this phenomenon is undoubtedly crucial.
Footnotes
Author Contributions
S.E.- Conceptualization, Formal Analysis, Writing- original draft, Writing- corrected draft; B.S.- Project administration, Investigation, Methodology, Resources, Software; T.S.S.- Project administration, Investigation, Methodology, Resources, Software; Y.A.H- Conceptualization, Supervision, Resources.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author.
