The earliest self-reflected mirror images of humans are likely to have been reflections in pools of still water, which is alluded to in the Greek myth of Narcissus falling in love with his own reflection. Manufactured mirrors may date back to approximately 6000 BC, made from naturally occurring volcanic glass. Therefore, over thousands of years, humans have become accustomed to seeing their own face as a mirror image reflected to them, rather than their real appearance that is observed by others. The increased use of smartphone cameras and selfie images over the past decade may have increased the tendency of people viewing their mirror images, often of their facial appearance.

The results of the present investigation found no significant gender differences between observers for any photographs except the frontal at rest photograph of the volunteer. For this individual’s set of photographs, the majority of observers preferred the original true image, with only a small number showing preference for the mirror image, and an even smaller number having the same preference. This is evidence that most people appear to prefer the true image of others, which is how they normally view them. Stage of clinical training for the observers did not appear to be a significant determining variable. The majority of observers preferred their own original photograph in frontal view at rest and smiling. Conversely, most observers preferred the mirror image of their selfie photograph at rest and while smiling, with only a minority having the same preference for both their own original and mirror image photographs. Differences in perception were also found between the observers’ own photographs in frontal view at rest and selfie at rest, and between preferences of photographs of frontal view while smiling and selfie while smiling. The observers appear to prefer their image the way they view themselves in a mirror. The selfie is how other people view an individual. If a selfie is flipped horizontally, that is how an individual sees themselves in a mirror. Therefore, it is logical that most observers preferred the mirror image of their selfie, which is how they would see themselves in a mirror.

The results of this investigation may be relevant to multiple specialties in medicine, surgery and the social sciences. For example, Felig and Goldenberg [3] investigated the phenomenon of selfie-evaluation and behavior. In an extensive meta-analysis of the available literature, they found inconsistent evidence in the relationship between selfie behaviors and appearance-specific self-evaluations. However, they did find that selfie editing was related to negative self-evaluations both generally and specific to appearance. It may be that time spent editing an image of oneself may be indicative of an underlying psychosocial issue, although this is purely speculative and requires further investigation.

Mombaerts and Missotten [4] described a case series of patients with an ocular prosthesis and their mirror images. They found that of the 16 patients with an ocular prosthesis, all preferred their mirror image. The authors acknowledged that family and friends often prefer the real image of the patient with an ocular prosthesis to the mirror image. De Runz et al. [5] looked at female plastic surgery patients, with particular emphasis on facial aesthetic surgery, in relation to preferences for standard photographs compared with mirror images. They acknowledged that familiarity with an image, e.g. a mirror image of themselves, may lead to patients preferring that image rather than their real image, as viewed by others or in a standard photograph. Faces are asymmetrical, so faces in photographs are different from those observed in mirrors. The main objective of this study was to assess whether female plastic surgery patients preferred standard photographs or mirror-reversed photographs of themselves. They found that of 214 patients, predominantly middle-aged, 73% preferred their mirror-reversed photographs, a phenomenon that was more pronounced among patients undergoing facial aesthetic surgery.

Pouwels et al. [6] looked at potential differences in aesthetic appreciation of a left peripheral facial palsy compared to a right peripheral facial palsy, in relation to true or mirror images. Pictures of patients with a facial palsy were reversed as a mirror image and shown as a pair together with the true image to 42 patients and 24 medical professionals familiar with facial palsy. The observers were asked to choose the most attractive photograph. One of the aims was to assess the preferences for the mirror or true images. When comparing mirror and true image, 90% of patients with a left-sided facial palsy chose their mirror image as most attractive, compared with only 30% of patients with a right-sided facial palsy. Medical professionals found a right-sided facial palsy cosmetically less attractive than left-sided. Patients, especially with a left-sided facial palsy, tended to choose their mirror image, although this choice seems to be influenced by hemispheric specialization and, interestingly, with familiarity. Vision is a physiological process. The eyes do not “see”, but detect light, allowing the perceiver’s brain to construct a visual representation of the viewed object. What we see is, to all intents and purposes, a mind-generated image of the real object. These results demonstrate that our understanding of how the brain analyses images, whether true or mirrored, requires further investigation.

Another area of clinical practice where distortions in photographs may be potentially significant is in relation to body image disorders [7]. Body image disorders, such as body dysmorphic disorder (BDD), and eating disorders related to disturbances in body image perception, such as anorexia nervosa and bulimia nervosa, have in common dissatisfaction with self-appearance, and a link between perception of appearance and feelings of low self-worth. The distortions in body image affect the individual’s self-esteem. Patients with BDD or the aforementioned eating disorders tend to spend an inordinate amount of time observing themselves in a mirror or in photographs [8], which in the age of smartphones is more likely to be with selfie photographs. Body image disturbances may elevate to delusionality, which may be related at least in part to perceptual aberrations and problems with visual processing [9]. It has been proposed that the negative body image perception in these conditions may be related to deficits in visual perception, with evidence particularly related to facial appearance, where it has been found that patients with BDD and anorexia nervosa demonstrate deficits in holistic visual processing (difficulty in seeing the big picture) but exaggerated visual processing in relation to small details [10]. Both groups have been assessed using functional magnetic resonance imaging and electroencephalography in relation to visual processing, and both have shown deficits in relation to facial visual processing [11]. One of the methods currently employed in cognitive behavior therapy for patients with BDD involves what is termed perceptual retraining, with which the patient looks at his or her mirror reflection, describing their image in a non-judgemental way. The intention is to draw attention away from obsessive attention to minor details in appearance [12]. As this therapy is related to mirror image assessments, it is important that evaluations related to assessments of mirror images compared to real images (e.g. standard photographs) are better understood [13, 14]. The potential relationship between visual processing disorders, body image disturbances, and self-evaluation of real and mirror images requires evaluation.

A potential limitation of this investigation was that the methodology was limited to static images, whereas it should be borne in mind that individuals observing themselves in a mirror may be in a dynamic state, e.g. smiling or possibly speech. During the dynamic process of speech, it has been observed that over three-quarters of people demonstrate greater amplitude of movement on one side of the mouth compared with the other [15]. Additionally, it would be useful to compare responses from a lay population, as the participants in this investigation were dental students.