An Eye-Tracking Investigation on Mirror Gazing: Those of higher self-esteem spent less time checking themselves in a mirror

Effects of Self-Esteem on Self-Viewing: An Eye-Tracking Investigation on Mirror Gazing. Jonas Potthoff, Anne Schienle. Behav. Sci. 2021, 11(12), 164. Nov 29 2021. https://doi.org/10.3390/bs11120164

Abstract: While some people enjoy looking at their faces in the mirror, others experience emotional distress. Despite these individual differences concerning self-viewing in the mirror, systematic investigations on this topic have not been conducted so far. The present eye-tracking study examined whether personality traits (self-esteem, narcissism propensity, self-disgust) are associated with gaze behavior (gaze duration, fixation count) during free mirror viewing of one’s face. Sixty-eight adults (mean age = 23.5 years; 39 females, 29 males) viewed their faces in the mirror and watched a video of an unknown person matched for gender and age (control condition) for 90 s each. The computed regression analysis showed that higher self-esteem was associated with a shorter gaze duration for both self-face and other-face. This effect may reflect a less critical evaluation of the faces.

Keywords: face; self-perception; mirror; eye-tracking; self-esteem; self-disgust; narcissism

4. Discussion

This study investigated whether gaze parameters while viewing one’s face in the mirror are associated with the personality traits of self-esteem, self-disgust, and narcissism propensity. High self-esteem indicates that, on the whole, individuals are satisfied with themselves and are not overly critical [30,31]. In previous eye-tracking research, elevated self-esteem was associated with attentional biases towards self-faces [13]. Surprisingly, in the present investigation, higher self-esteem was associated with shorter—possibly less critical—viewing of the own face. It is possible that people with high self-esteem need less time to evaluate themselves (critically), while low self-esteem seems to be associated with a more thorough and more prolonged evaluation of one’s facial appearance.

Following this interpretation, the shorter viewing time of the other face associated with higher self-esteem would imply that individuals with high self-esteem are also less critical of others [32]. It is possible that in previous research [13] this viewing pattern could not be identified due to brief exposure times. The participants looked at the faces until they identified them. This took them less than one second on average. Additionally, in previous research, faces were not viewed freely, and task demands (face identification) may have overruled a critical and thorough evaluation of the faces [13]. Future research needs to investigate the effect of exposure time and viewing tasks on gaze behavior concerning self-face and other-face.

Self-esteem did neither predict the number of fixations on self-face nor other-face. Many short fixations (i.e., hyperscanning) are commonly conducted when affective stimuli with negative valence and high intensity are explored [22,33]. In the present study, dynamic faces with neutral expressions were viewed, which do not induce high arousal, at least in non-clinical samples [19,20,21]. While self-esteem predicted a more thorough evaluation of faces (longer viewing time), it was not associated with visual hyperscanning, which would have required elevated arousal. Taken together, the results suggest that low self-esteem is related to a thorough but calm evaluation of faces.

Earlier investigations indicated that the number of fixations on specific face areas differs between self-face and other-face viewing. Hoffmann et al. [8] observed that more fixations were conducted on the eyes when viewing unfamiliar faces than the own face. We refrained from using a similar analysis approach (an analysis of fixation counts for specific facial areas) because of insufficient spatial precision of the measuring procedure. There are slight interindividual differences in the position of facial features (e.g., location of eyes, mouth), which cannot be detected with sufficient spatial accuracy during mirror viewing. It would have been possible to control for the location of the eyes during mirror viewing by adjusting the chinrest for each participant until they see their eyes at a predefined mirror location. This procedure would, however, lead to self-face exposure before the eye-tracking experiment begins. Therefore, we focused on the gaze behavior for the face as a whole.

There was no relationship between NPI scores and gaze behavior. Narcissism propensity was neither associated with a more thorough (prolonged) viewing nor hyperscanning. These null-findings contrast our expectations and previous research, especially for self-face viewing [15]. In the present paradigm, participants viewed self-faces and other faces one after another and not simultaneously (e.g., as image pairs). It is possible that associations between narcissism propensity and viewing behavior did not show up because participants could not directly compare their faces with the faces of others. Furthermore, participants did not perform any task in which they could have compared their performance with the performance of others. Future studies should investigate whether narcissism propensity is associated with specific self-face viewing behaviors in competitive contexts (i.e., when comparing one’s own and others’ physical appearance).

The NPI assesses feelings of self-importance. However, narcissism has different facets. It has been suggested to distinguish between grandiose and vulnerable narcissism [34,35]. Vulnerable narcissistic individuals are anxious, defensive, and avoidant, while grandiose narcists are extraverted and self-satisfied [36]. Additionally, there are intrapersonal and interpersonal aspects of narcissism [37]. To maintain a grandiose self, people with a high narcissism propensity might, on the one hand, strive for feelings of uniqueness (intrapersonal) [38]. On the other hand, they can also devaluate others or strive for supremacy (interpersonal) [39]. Therefore, more differentiated NP measures should be used in the future, which can differentiate between these separable yet related expressions of narcissism that should prompt different styles of self-viewing [40].

Self-disgust also showed no association with viewing the own face. Self-disgust has been conceptualized as a dysfunctional personality trait [12,41] which typically has a very low level in mentally healthy individuals [41]. This was also the case in the present investigation. In contrast, Ypsilanti, Robson et al. [6] analyzed eye-tracking data from participants with mental health problems. This group with elevated self-disgust displayed avoidance of self-viewing. Therefore, the present null results do not contrast previous findings [5,6]. Future research should further investigate self-face viewing in conditions with elevated self-disgust, such as depression [2,42], disordered eating [43,44,45], and body dissatisfaction [45,46,47]. For example, body dysmorphic disorder (BDD) is associated with high self-disgust [47] but not with avoidance of self-viewing. Many patients with BDD even report spending a significant amount of time checking themselves in the mirror [45]. Mirror viewing paradigms can therefore contribute to a deeper understanding of self-viewing avoidance as well as excessive mirror viewing in eating disorders and BDD. For these disorders, mirror exposure is a commonly used therapy component [48].

We also investigated possible gender effects on face-viewing behavior. In general, males, as well as females, spent the majority of the 90 s-viewing time on the faces. Thus, there was no indication of face avoidance. Male participants were characterized by a longer total viewing time. They showed a longer gaze duration for both the own face and the other face. Previous studies have already demonstrated that face exploration dynamics (for other faces) differentiate men and women [49,50]. In an investigation by Coutrot et al. [49], the participants (203 males and 202 females) watched videos of actors. Male participants showed longer fixation durations for the faces than female participants. However, this difference was small. Overall, face viewing has many functions and serves multiple purposes (e.g., aiding speech perception [51], emotion recognition [52], person identification [53], or affective evaluation [15]). Moreover, context factors (e.g., cultural, social) have to be considered as well [54]. It needs further investigation whether differences in face-viewing intentions might have contributed to the observed gender difference in gaze duration.

It is plausible that several limitations might have influenced the results obtained. We studied a convenience sample, and the participants were healthy, predominantly young students. Therefore, the present results cannot be generalized to other samples. Furthermore, the study did not differentiate between different aspects of narcissism propensity.