Behaviours Related to Selfie Addiction, Narcissistic Traits and Social Validation Among Young Mexicans

1. Introduction

Generally speaking, social media platforms are currently very popular applications where people interact socially; in this context, selfies are extremely common; furthermore, social media is a key channel for personal expression and communication in many people’s lives (Fox & Rooney, 2015; Sakthivel & Thangamuthu, 2019; Yang & Crespi, 2025). From this perspective, selfies are a means of communication, such that both the message and the choice of medium through which it is conveyed depend on those who are communicating; therefore, it is worth clarifying that photographs one takes of oneself are known as selfies; there are times when other people appear in them. Furthermore, a selfie consists of three elements: the display of the person’s face, the act of taking and disseminating the photograph, and the purpose of sharing it (Akinola et al., 2022; Walsh & Baker, 2017).

According to this line of reasoning, people may engage in social interaction to gain, maintain or lose specific rewards by taking and posting selfies, which may cause them to experience certain emotional states; in particular, adolescents who receive a more significant number of positive responses to the selfies they post will spend more time checking their social media accounts searching for comments and ‘likes’ (Monacis et al., 2020; Zeng et al., 2024). Furthermore, studies highlight that people who take selfies have a greater perception of being accepted, valued or recognized by others; they also perceive themselves as more competent in terms of their abilities, achievements, and performance; and they feel they have greater control over their lives (Corsbie-Massay & Sargent McLaren, 2024). In line with this approach, various studies have noted that adolescents use Instagram to watch silly, light-hearted videos and to share their selfies via photos and videos to attract more attention from their peers; furthermore, other studies have shown that comparisons relating to appearance, as well as interaction on Instagram, were significantly associated with body dissatisfaction in both women and men (Legkauskas & Kudlaitė, 2022; Mustafa et al., 2021).

In this regard, it is worth noting that posting more selfies more frequently is associated with the number of followers each user has; furthermore, when women frequently post their photographs on Instagram, this can lead them to judge themselves and observe others (Barry et al., 2017; Chang et al., 2019). Along the same lines, Ridgway & Clayton (2016) have noted that body image satisfaction is associated with an increase in the number of selfies posted on Instagram. Furthermore, research such as that by Verrastro et al. (2020) suggests that some Instagram users edit their photos to conform to the beauty standards of this social media platform; these individuals express a greater fear of their appearance being negatively evaluated and also display signs of anxiety related to their body image. Similarly, an increase in body image issues is associated with an exaggerated portrayal of thinness and beauty resulting from frequent exposure on Facebook, TikTok, and Instagram; in addition to this, there are studies providing evidence of detrimental effects on the mood and body satisfaction of young adults who are exposed, in the short term, to aesthetic stereotypes on Instagram (Castellanos Silva & Steins, 2023; Goyal & Gautam, 2024).

Within this framework, narcissism is characterized as a complex personality trait comprising a lack of empathy, a sense of superiority and grandiosity; as well as an exaggerated self-image; furthermore, it is also important to take current perspectives into account to better interpret narcissism (Gnambs & Appel, 2018; Runcan et al., 2023). From a contemporary perspective, recent research into narcissism has incorporated new dimensions of grandiosity, namely individuals who resort to techniques to belittle others, control their ego and elevate their worth; whilst reactive anger, distrust, psychological distress, hostility, negative affect, egocentrism, mistrust, and low self-esteem are traits of vulnerable narcissism (Prabhu et al., 2024; Weiss et al., 2019). Along the same lines, people with vulnerable narcissistic traits tend to counteract and acknowledge the challenges posed to their self-image; in similar terms, vulnerable narcissism is associated with depression, isolation, psychological distress, anxiety, feelings of inferiority, as well as hypersensitivity to criticism; from a similar perspective, grandiose narcissism is associated with characteristics such as a lack of modesty, egocentricity, interpersonal dominance, manipulation, insensitivity, exhibitionism, and a need for recognition from others (Kaufman et al., 2020; Prabhu et al., 2024).

Based on this conceptualization, the literature has also indicated that users who frequently post selfies and display a strong intention to do so appear to exhibit narcissistic traits; in light of this, it is necessary to explore in greater depth how digital environments enable individuals with narcissistic traits to manage and maintain control over the information and activities they share; Furthermore, this evidence suggests that individuals with high scores on narcissism scales reported frequent use of social media platforms such as Twitter, Facebook, and Instagram (Bergman et al., 2011; Martingano et al., 2022; Sung et al., 2016). More specifically, some research indicates that narcissism is associated with selfies; however, no such link is found for photos with friends or activities centred on others (Bergman et al., 2011). In line with this approach, it is worth noting that people with narcissistic behaviours are more likely to use social media more frequently; following this logic, research has revealed that adolescents with higher levels of narcissism experience greater problems regarding social media use; Similarly, studies have found that selfies serve as a psychological mechanism of particular interest in the pursuit of attention (Darwaish & Nazneen, 2022; Jiang et al., 2023; Sung et al., 2016).

In line with these arguments, research indicates that people select and post their best selfies on social media to present themselves as extraordinary and different; it has also been documented that one of the reasons people post selfies on social media is to gain acceptance or admiration as feedback through ‘likes’ and ‘comments’ (Srivastava et al., 2018; Vijayan et al., 2024). Furthermore, social media use is linked to certain personality traits and narcissism, and there is evidence that self-esteem, narcissism and social media are interrelated; it has also recently been documented that narcissistic rivalry and admiration are significantly and positively associated with problematic social media use (Asad et al., 2022; Mosanya et al., 2024; Usman et al., 2024). Moreover, people who tend to take and share selfies—a behaviour encouraged by social media and which promotes self-presentation—are those with higher levels of narcissism and envy; at the same time, selfie addiction is significantly associated with the desire to be seen on social media among young adults; from a similar perspective, research describes how perfectionist and idealistic attitudes towards others and oneself are linked to self-promotional behaviour on social media and narcissism (Abbasi et al., 2024; Baran & Karaca, 2024; Sánchez-Domínguez et al., 2026). Following this line of reasoning, there are also studies indicating that people who display narcissistic traits tend to take an excessive number of selfies (Puthiyakath et al., 2024).

Finally, other studies have shown that problematic use of social media is positively correlated with grandiose and vulnerable narcissism; from a similar perspective, there are studies indicating that certain characteristics of social media affect self-esteem and body objectification, primarily in men, as evidenced by beliefs regarding body control, appearance-oriented habits, body-shame, and photo manipulation practices, all of which are associated with how people value their bodies; In relation to this point, research indicates that the intensification of the media ideal can lead to increased feelings of body-shame and appearance comparison among female Instagram users (Boursier & Gioia, 2022; Galli et al., 2025; Schettino et al., 2023).

In this context, studies should investigate the dynamics and consequences of social media to understand and manage its negative effects; along the same lines, the available literature suggests that further research is needed into psychological factors such as personality traits, cognition, motivations, and attitudes that may influence the onset, development, and persistence of ‘selfitis’; furthermore, evidence from research with adolescents has shown that problematic social media use can be categorized into distinct risk profiles, indicating the need to analyse specific indicators rather than merely overall scores for problematic use (Balakrishnan & Griffiths, 2018; Li et al., 2020; Nene & Olayemi, 2025). Complementarily, current studies indicate that adolescents are at risk of social media addiction, which is associated with socio-emotional and mental health processes, thereby increasing the need to examine psychological variables related to social media use (Accorinti et al., 2025). In light of this, it is necessary to use validated scales on selfie addiction to better describe psychological patterns such as objectification, narcissism, body image concerns and social media engagement among those who take selfies selectively and those who do not (Corsbie-Massay & Sargent McLaren, 2024). Furthermore, there remains a lack of evidence regarding the interaction between self-promotional behaviour and perfectionism—which may be associated with narcissism—and how individuals may manipulate their appearance on social media; moreover, this issue is exacerbated by the fact that it is essential to explore the effects of social media on psychological processes (Abbasi et al., 2024; Mosanya et al., 2024). Given this research gap, there is a need to investigate the underlying psychological mechanisms in young adults about social media addiction; furthermore, the phenomenon is relevant because it is necessary to study key psychological factors such as emotional regulation, insecurity, cognitive styles, social anxiety, impulsivity, and concern about one’s personal image on social media; furthermore, it is necessary to analyse in depth the possible association between selfie addiction and the motives behind it (Fam, 2024; Galli et al., 2025; Usman et al., 2024). Finally, research into the possible relationships between the rewards offered by Instagram is limited (Weiland & McLoughlin, 2025).

The aim of this study is therefore to predict selfie-related behaviours among students at the Autonomous University of Zacatecas (UAZ, for its acronym in Spanish), Mexico. This will be achieved using an adapted version of the NPI-40 by Raskin & Terry (1988), the Psychometric Scale of Selfie Addiction (PSSA) by Arumugam & Nagalingam (2015), as well as the indicators of the need for social validation proposed by Fegan & Bland (2021). All of the above was analysed using exploratory factor analysis and hierarchical linear regression.

In line with the established objective, this study hypothesised that narcissistic traits would be associated with greater engagement in selfie-addiction behaviours. This is primarily driven by the need for social validation via social media. Consequently, it is proposed that narcissism be positively associated with behaviours related to selfie addiction and, in turn, that both are linked to indicators of the need for social validation on social media. However, unlike previous studies which focused primarily on factor structure analysis and the general relationships between selfie addiction and narcissism, this study assesses the incremental predictive value of specific social validation variables on Instagram, after controlling for the narcissistic facets derived from a fitted structure of the NPI-40.

2. Materials and Methods

2.1. Design

The research was designed as an empirical, non-experimental, cross-sectional and correlational-predictive study. The criterion variable (CV) was the mean score of the PSSA items, which was termed the PSSA total (behaviours related to selfie addiction). To operationalize the need for social validation on social media, the six predictor variables proposed by Fegan & Bland (2021) were used. Under this approach, narcissistic traits were measured using an adapted version of the NPI-40 by Raskin & Terry (1988).

2.2. Participants

Initially, 265 students from the UAZ, aged between 13 and 35, selected through convenience sampling, took part. They were asked to provide sociodemographic data regarding their sexual orientation, categorized into five responses: heterosexual, bisexual, demisexual, pansexual, and homosexual. It was decided to exclude cases that did not fall within the 14 to 23 age range, to limit the study to students at upper-secondary and undergraduate levels, in line with the methodological criteria used in previous studies (Abbouyi et al., 2026). Similarly, a further five responses were excluded as they duplicated the level of study the participants were currently undertaking. Consequently, a sample of 255 students remained.

2.3. Instruments

2.3.1. Adjusted NPI-40

Given the exploratory nature of the research, an adapted version of the NPI-40 instrument by Raskin & Terry (1988) was used. After adjusting the NPI-40, eighteen items were selected to measure narcissistic traits. The items had the following response options: 1 = Strongly disagree, 2 = Disagree, 3 = Neither agree nor disagree, 4 = Agree and 5 = Strongly agree, as in previous studies (Barelds & Dijkstra, 2010; Gioia et al., 2023; Huang et al., 2023; Velji & Schermer, 2024). In turn, the Adjusted NPI-40 (NPI-40 A) groups together items that focus on agentic grandiosity, authority, exploitation, exhibitionism, and vanity.

2.3.2. PSSA Total

The ten items of the PSSA developed by Arumugam & Nagalingam (2015) were used to measure behaviours related to selfie addiction. Similarly, the items were scored on a scale of 1 = Strongly disagree, 2 = Disagree, 3 = Neither agree nor disagree, 4 = Agree and 5 = Strongly agree, as in previous studies (Barelds & Dijkstra, 2010; Gioia et al., 2023; Huang et al., 2023; Zeng et al., 2024). Collectively, they measure self-perceived behavioural priority given to taking selfies in everyday life; the presence of social impairment and disengagement from context; furthermore, the excessive compulsion to take selfies; in addition, taking selfies as a form of emotional regulation; likewise, the emotional distress caused by taking selfies; and finally, the self-perception of needing help with an addiction to taking selfies.

2.3.3. Predictive Factors

The six variables proposed by Fegan & Bland (2021) were used as predictors of the need for social validation on social media; the items are as follows: PRE_1 “The average amount of time I spend interacting on Instagram”, PRE_2 “How often I post on my Instagram account”; both statements have response options 1=Never, 2=Every few weeks, 3=A few times a week, 4=Once a day and 5=Several times a day, as recommended by Fegan & Bland (2021). The items PRE_3 “I care about the number of ‘likes’ or comments I receive on my posts and feel good about myself when this number is high”, PRE_4 “I feel annoyed or jealous when I see other people posting positive updates about their lives and find myself comparing my life to theirs”, PRE_5 “A negative comment on my post would make me feel bad and ruin my day” and PRE_6 “My image on Instagram is different from who I am in real life”, also proposed by Fegan & Bland (2021). These items capture the time spent interacting on Instagram, the frequency of posting on Instagram, sensitivity to negative feedback in the comments, social comparison, and a different self-presentation on Instagram.

2.4. Procedure

Based on the above, all items from the NPI-40 and the PSSA, as well as the six predictor variables, were initially entered into Google Forms. Data collection took place between September and November 2024 among students at the Autonomous University of Zacatecas (UAZ), using convenience sampling. An invitation was sent to them by lecturers and programme coordinators to complete the Google Forms link via institutional communication channels, such as WhatsApp groups, as had been done in previous studies (Asad et al., 2022; Legkauskas & Kudlaitė, 2022; Vijayan et al., 2024). The WhatsApp message included an explanation of the research, eligibility criteria and ethical considerations. The first screen of Google Forms explained to participants the objectives of the study, the voluntary nature of participation, their right to withdraw from the survey at any time without facing any consequences, and the confidentiality of their data. For students under the age of 18, informed consent from a parent or legal guardian was required, in addition to the minor’s own assent. Furthermore, no personal data that directly identified the respondent was collected; therefore, the responses were analysed anonymously and exclusively for academic purposes. The inclusion criterion was being enrolled at the UAZ at the time, as well as providing informed consent; in the case of minors, written consent from a legal guardian was also required, as suggested in previous studies (Andrade et al., 2023; Basith et al., 2023; Gioia et al., 2023; Huang et al., 2023; Toubasi et al., 2023). It is important to note that the research project was approved by the UAZ Higher Education Teaching Unit Council at its meeting on 7 May 2024; this Council acts as the Ethics Committee under UAZ regulations, reviewing the relevance of the research project and ensuring that respondents’ data were handled confidentially. Similarly, the project was registered with the UAZ Research and Postgraduate Coordination Office under registration number UAZ-2024-39072, as has been the case in previous studies (Galli et al., 2025; Skowronski et al., 2021).

2.5. Statistical Analysis

The statistical analyses were carried out in R 4.5.2; a descriptive analysis of the research items was then conducted. At a later stage, an exploratory factor analysis (EFA) was performed on the polychoric correlation matrix using the minimum residuals extraction method and oblimin oblique rotation. Once this had been completed, a parallel analysis was carried out, which suggested three factors and two observed composites. Factor loadings ≥ 0.30 were retained, whereas those below this threshold were removed to identify a more parsimonious structure, in accordance with previously validated protocols (Corsbie-Massay & Sargent McLaren, 2024; Womick et al., 2020). Consequently, after running the EFA to achieve a more efficient internal factor structure, 18 items from the original NPI-40 were retained; it was therefore decided to name this version the Adjusted NPI-40 (NPI-40 A). Furthermore, the solution showed an RMSEA ≈ 0.100, TLI ≈ 0.845, and RMSR ≈ 0.022, suggesting a mixed model fit; furthermore, the preliminary EFA structure explained approximately 63.8% of the variance (Brown, 2015; Costello & Osborne, 2005; Juárez Velarde & Valdez, 2024; Tanzilli et al., 2016; Thurstone, 1947). Subsequently, statistical analyses were carried out on the 18 retained items, including Cronbach’s alpha (α), McDonald’s omega (ω) and the Kaiser-Meyer-Olkin (KMO) statistic for each factor, in accordance with the procedures applied in similar studies (Balakrishnan & Griffiths, 2018; Corsbie-Massay & Sargent McLaren, 2024; Dunn et al., 2014; Jiang et al., 2023; Runcan et al., 2023; Sánchez-Domínguez et al., 2026; Verma et al., 2025). Factor 1 (Fc1) had an α = 0.782, an ω = 0.843 and a KMO = 0.796; Factor 2 (Fc2) had an α = 0.782, an ω = 0.829 and a KMO = 0.796; Factor 3 (Fc3) had an α = 0.844, an ω = 0.878 and a KMO = 0.732; in the case of Factor 4 (Fc4), which will be referred to as the Observed Composite (OC) comprising two items, reliability was calculated using the Spearman-Brown (SB) test, which yielded a value of 0.914; and for Factor 5 (Fc5), which also consists of two items and is likewise considered an observed composite, the reliability was SB = 0.843. Similarly, to verify the overall suitability of the exploratory factor analysis (EFA), Bartlett’s sphericity test was conducted, yielding p ≤ 0.001 and the Kaiser-Meyer-Olkin (KMO) statistic was calculated, yielding a value of 0.722, in line with previously reported methods (Gioia et al., 2023; Juárez Velarde and Valdez, 2024).

To analyse the predictive power of the need for social validation and narcissistic traits regarding behaviours related to ‘selfie’ addiction as measured by the PSSA total score, a hierarchical linear regression analysis was conducted, following the procedure used in previous studies (Cary et al., 2024; Kwolek, 2012; Legkauskas & Kudlaitė, 2022; Martingano et al., 2022; Usman et al., 2024). Model 1 (M₁) was estimated using age; Model 2 (M₂) incorporated factors derived from the EFA; finally, Model 3 (M₃) included the social validation items proposed by Fegan & Bland (2021). Furthermore, standardized and unstandardized coefficients, 95% confidence intervals (CI), standard errors, and the model indicators R, R², adjusted R², RMSE, AIC and BIC were reported. A comparison between models was also reported using the F-test for change and the change in R². Finally, multicollinearity was assessed using tolerance and the variance inflation factor (VIF), in accordance with the methodological criteria used in previous studies (Kwolek, 2012).

3. Results

Among the participants with valid data (n = 255), the mean age was 17.65 years with a SD = 2.48; the Total PSSA score was M = 1.83 with a SD = 0.70; the facets of the NPI-40 A that showed the highest scores were vanity, with a mean of 3.94 and a standard deviation of 0.64; exploitation, with a mean of 3.42 and a standard deviation of 1.10; and agentic grandiosity, with a mean of 3.29 and a standard deviation of 0.70. Meanwhile, among the social validation indicators, PRE_1 obtained the highest mean (M = 4.11) and a standard deviation (SD) of 1.20, whilst PRE_4 obtained the lowest mean (M = 1.80) and a standard deviation (SD) of 0.98.

The ‘agent-based grandiosity’ factor (Fc1) comprised items NPI_14, NPI_15, NPI_37, NPI_3, NPI_18 and NPI_38, with loadings ranging from 0.80 to 0.46 (Table 1). In addition, Fc2, which comprises items NPI_1, NPI_8, NPI_4, NPI_2 and NPI_7—which represent narcissistic traits of authority—had loadings ranging from 0.83 to 0.42 (Table 1). Similarly, the items NPI_28, NPI_29 and NPI_30, which make up Fc3 (exploitation), have loadings ranging from 0.86 to 0.74 (Table 1). This appears to reflect a tendency to use manipulation to gain personal benefits. The composite observed in Fc4 (exhibitionism), comprising NPI_23 and NPI_25, has loadings of 1.00 and 0.86 respectively (Table 1). Finally, Fc5 (vanity), which also comprises only two items, is an observed composite consisting of NPI_32 and NPI_33, with loadings ranging from 0.90 to 0.78 (Table 1). This can be explained as a narcissistic facet characterized by a desire to receive special or preferential treatment.

Considering the above, a hierarchical linear regression was carried out using the mean total PSSA score as the criterion variable. Model 2 (M₂), which includes age as part of the analysis, indicates a strong association with an R value of 0.376, as well as an R² value of 0.141 and adjusted R² of 0.120, suggesting that age and narcissism factors explain 14.1 per cent of the variance in behaviours related to selfie addiction (Table 2). Furthermore, p ≤ 0.001 (Table 2). The results suggest that age showed an initial relationship with behaviours related to selfie addiction, given that the respondents’ ages ranged from 14 to 23 years; it was therefore to be expected that there would be differences between adolescence and young adulthood. In turn, behaviours related to selfie addiction are not only associated with age, but also with factors such as agentic grandiosity, authority, exploitation, exhibitionism and vanity, such that behaviours related to selfie addiction represent an evolution in the development of identity and emotional regulation, as well as being sensitive to narcissistic facets.

In light of the above, it should be added that Model 3 (M₃) reportedly includes the dimensions PRE_1, PRE_2, PRE_3, PRE_4, PRE_5 and PRE_6 proposed by Fegan & Bland (2021). This is indicated by the results R = 0.543, R² = 0.295 and an adjusted R² = 0.260, which show that M₃ explains 29.5 per cent of the variance (R²), and when adjusted, indicates 26.0 per cent of the explained variance (adjusted R²) (Table 2). Furthermore, an increase in R² of 0.154 is observed, representing an additional 15.4 per cent. This corresponds to a moderate increase in F(6, 241) = 8.748, p ≤ .001 (Table 2). Furthermore, it is important to note that the root-mean-square error (RMSE) also decreases moderately from 6.459 to 5.853 (Table 2).

The M₁ indicates that at younger ages there is an initial association with behaviours related to selfie addiction (PSSA total); this is because age yields a β value of -0.129, along with a t-coefficient of -2.062 and a p-value of 0.04; a 95% confidence interval (CI) with a lower limit of -0.12 and upper limit of -0.003 (Table 3). However, in model M₂, age is no longer a statistically significant variable when the narcissistic trait factors indicated by the EFA are incorporated, as the p-value is 0.28 (Table 3). In contrast, when the narcissistic trait factors were added to M₂, Fc4-CO yielded results of β=0.306, along with a t-coefficient of 4.545 and a p-value of 0.001; a 95% confidence interval (CI) with a lower limit of 0.21 and upper limit of 0.531 (Table 3). In turn, Fc3 showed a β value of 0.201, a t-coefficient of 3.142 and a p-value of ≤ 0.002; a 95% confidence interval (CI) with a lower limit of 0.081 and upper limit of 0.353 (Table 3).

In light of the above, the inclusion of the predictor variables PRE_1, PRE_2, PRE_3, PRE_4, PRE_5 and PRE_6 yielded the relevant results set out below. Consequently, PRE_2 had an effect on M₃ with β = 0.289, t = 4.618 and p = 0.001 with a 95% CI with a lower limit of 0.213 and an upper limit of 0.530 (Table 3). It is worth noting that age only showed a negative effect on M₁, but, as with M₂, it lost significance in M₃ when controlling for narcissistic trait factors and items relating to the need for social validation. It is important to note that there is no multicollinearity, as indicated by the variance inflation factor (VIF), which ranges from 1.047 to 1.764, and a high tolerance of approximately 0.567 to 0.955 (Table 3).

4. Discussion

Our findings provide preliminary exploratory evidence that the NPI-40 A, comprising three factors and two observed composites, exhibits a moderate overall fit. Furthermore, Total PSSA—which represents behaviours related to selfie addiction—is primarily associated with the narcissistic facets of exploitation and exhibitionism, as well as an indicator of the frequency of Instagram posts, and greater sensitivity to negative comments on social media posts. Our results support previous studies such as that by Casale & Banchi (2020), who, in their systematic review, found evidence of the need for self-reinforcement, as narcissism influences the production of selfies and this, in turn, increases the levels of narcissism reported by social media users over time.

Although the EFA was interpretable and relatively parsimonious, the overall indices of RMSEA ≈ 0.10 and TLI ≈ 0.845 should be interpreted with caution, as they indicate a moderate fit. These results should therefore be regarded as preliminary evidence. However, we consider the EFA to be appropriate given the KMO result of 0.722, as well as the fact that Bartlett’s sphericity test indicated p ≤ 0.001. Furthermore, the parallel factor analysis suggested retaining 18 items from the NPI-40, distributed across three factors and two observed composites with loadings greater than 0.30 in the dimensions of agentic grandiosity, authority, exploitation, exhibitionism, and vanity. The factors of agentic grandiosity, authority, and exploitation showed Cronbach’s alpha and McDonald’s omega coefficients ranging from acceptable to good. Meanwhile, the observed composites of exhibitionism and vanity demonstrated good Spearman-Brown reliability; however, they should not be interpreted as having factor status. Furthermore, even with the reduction in the number of items, the EFA retained conceptually defensible dimensions of narcissistic facets such as agentic grandiosity, authority, exploitation, and two observed composites of exhibitionism and vanity. It is therefore important to bear in mind that the structure is preliminary and awaits confirmatory validation in independent samples.

Whilst bearing these methodological considerations in mind, exhibitionism, and exploitation are of particular theoretical interest, as they emerge as particularly relevant dimensions. Exploitation and exhibitionism were statistically significant predictors of behaviours related to selfie addiction, as measured by the total PSSA; this finding suggests that behaviours related to selfie addiction may be associated with narcissistic facets of exploitation and exhibitionism via social media, which could be linked to a context of self-presentation in a controlled manner. Therefore, our results provide further evidence to support studies such as that by Arpaci et al. (2023), who indicated in their findings that exhibitionism acts as a moderator in the association between the behaviour of posting selfies and psychological needs. Similarly, our research is consistent with that of Boursier et al. (2020), who demonstrated in their study that positive expectations regarding selfies are correlated with grandiose or vulnerable narcissistic facets in both women and men.

Furthermore, the increase in R² = 0.154 showed that adding the predictor variables to M₃ not only provided a predictive indicator of the possible presence of narcissistic traits of exhibitionism – and that these are not isolated; but also that the PRE_2 item, which measures the frequency of posting on Instagram, captures situations associated with higher scores on behaviours related to selfie addiction. We consider that our results provide further evidence that the need for social validation is reflected in behaviours related to selfie addiction. Furthermore, they are consistent with previous studies whose findings indicate that users who spend more time on Instagram are those who report higher levels of body dissatisfaction, have low self-esteem and compare their physical appearance with that of other users; similarly, young people view social media as a tool for gaining recognition through ‘comments’ and ‘likes’, as well as for demonstrating their worth; the more ‘likes’ they receive, the more evident it is that they have made a name for themselves and are popular, so that many people recognize and see them (Alfonso-Fuertes et al., 2023; Nguyen et al., 2025). Similarly, our findings appear to be in line with those of Chen (2025), who notes that adolescents’ self-esteem improves when they receive positive comments and fewer negative ones on social media.

It is worth noting that not all narcissistic facets are linked to selfie-addictive behaviour. Similarly, the traits of agentic grandiosity, authority, and vanity were not significant in this study. By contrast, exhibitionism, and exploitation did show significant associations. About the indicators of social validation relating to the frequency of Instagram posts, as measured by PRE_2, and PRE_5, which relates to sensitivity to negative comments, these were indeed statistically significant predictors. In contrast, the predictor items PRE_1, PRE_3, PRE_4 and PRE_6 were not significant. It is also important to note that individual items were used as predictors, which may have affected the accuracy and attenuated certain effects. Therefore, these results should be interpreted with caution, given that some studies mention the existence of optimized versions of the real self and the self-image; as the aim is to present an artificially optimized image, this could negatively influence one’s self-perception of one’s own attractiveness (Ozimek et al., 2023). It is also worth noting that age proved to be an unstable predictor; this is because M₁ indicated that younger age is associated with greater levels of behaviour related to ‘selfie addiction’; however, when items measuring the need for social validation and narcissistic facet factors were included, the effect of age disappeared.

With due caution, at a theoretical level the results indicate that the association between behaviours related to selfie addiction and facets of narcissism is neither universal nor homogeneous; therefore, narcissistic traits should not be analysed as a single block, but rather by facet. It is also worth noting that the narcissistic traits that carried the most weight were exhibitionism and exploitation through selfies; furthermore, the need for social validation does not act as a single block either, but in our research, the frequency with which users post on Instagram, as well as their sensitivity to receiving negative comments on their posts, stand out more than other variables.

4.1. Implications for Psychological Health and Digital Education

In practical terms, our findings suggest that prevention work with upper secondary and undergraduates should not focus solely on the amount of time spent on social media. It is even more important to consider psychoeducational interventions that focus on students’ emotional investment in how they present themselves on social media, the frequency of their Instagram posts, and their reactions to negative comments they receive on social media. In line with this argument, emotional regulation, digital literacy and psychoeducational support can help to reduce problematic patterns related to the search for social validation on social media.

It is worth noting that, from a methodological perspective, it may be more useful to investigate narcissism by its various dimensions rather than using a total score. Similarly, it would be advisable to measure the need for social validation using multi-item scales, as these offer greater precision than isolated statements. Furthermore, the NPI-40 A may be a useful exploratory tool; however, it is necessary to carry out a confirmatory factor analysis (CFA) and replicate the findings in independent samples.

Despite these findings, we are aware that our research may have limitations. The first is that self-report measures were used, and the study focused on Instagram; therefore, the results cannot be generalized to other contexts and social media platforms. Furthermore, the study employed convenience sampling and a cross-sectional design; consequently, a causal relationship cannot be established. Additionally, the sample was drawn from a single university, and the age range of the respondents was between 14 and 23 years. Finally, the predictors of the need for social validation are individual items, and there are two composite measures, each comprising only two statements. Therefore, future research might consider distinguishing more clearly between the frequency of posting selfies and the general frequency of posts, as well as the editing of photographs and selfies, and the pursuit of social approval.

5. Conclusions

In summary, the findings provide preliminary exploratory evidence regarding an adjusted version of the NPI-40 and suggest that behaviours associated with ‘selfie addiction’ are linked in distinct ways to certain narcissistic traits and to specific indicators of the need for social validation on Instagram. In particular, exploitation, exhibitionism, sensitivity to negative comments and posting frequency on Instagram were, statistically speaking, the most significant predictors in the final model. It should be noted that the relationship between behaviours associated with selfie addiction and facets of narcissism was not uniform, nor did the need for social validation act as a single block. An interpretable factorial solution comprising three factors and two observed composites was obtained; however, confirmatory validation in independent samples should be considered.

Furthermore, behaviours associated with selfie addiction are linked to narcissistic traits of exhibitionism and exploitation. Consequently, without considering the behaviours associated with selfie addiction and the need for social validation, narcissism cannot be understood in its entirety. Furthermore, the findings suggest the relevance of implementing psychoeducational strategies that focus on the frequency of Instagram posts, emotional regulation and high sensitivity to negative comments on social media posts. However, it is important to note that the nature of the research is cross-sectional, the sampling was convenience-based, the survey is self-reported, and the preliminary nature of the NPI-40 A structure mean that the results must be interpreted with caution and tested in future research.