1. Introduction
Research on interhuman eye contact has been conducted for millennia in populations with and without psychiatric disorders. So-called “atypical” eye contact is considered a core symptom of some psychiatric disorders. However, the question arises of what can be understood as “normal” eye contact. Eye contact is commonly defined as “a direct look exchanged between two people” [
1], yet this explanation leaves much room for interpretation. For example, a direct look is described as a gaze that is a maximum of 37 degrees to the left or right of the other person’s eyes [
2], while other research has shown that the experience of a direct look depends on the head direction [
3]. It is not clear whether the direction of a gaze is determined only by the position of the pupil relative to the sclera or by the entire eye region, including the eyebrows. Studies have suggested that eye contact only exists when there is also “you-awareness” of the other during the exchange of a direct look [
4], that eye contact is a form of mutual transaction [
5] or that eye contact implies connectedness [
6]. Because of these different approaches to eye contact, the concept may be applied differently, which could impede transferability between studies. To ensure transparency, the current review uses eye contact in the broadest sense: as a mutual gaze between two people.
This mutual gaze is considered fundamental to human communication and is possibly its most important aspect [
7,
8,
9,
10]. Human eyes convey a variety of complex social and emotional information [
11], and it has been found that the receiver’s perception of mental states is based on expressive eye behavior. It is not only the receiver’s perception that seems to play a role in this process, but also, for example, the viewing direction of the eyes [
12]. A contribution to this gaze discrimination was the evolutionary adaptation of depigmentation of the human sclera, in which the dark iris and gaze direction became clearer, allowing further optimization of eye contact in humans [
13]. Other influences on how eye contact is interpreted include pupil dilation and constriction [
14]; gaze cuing [
15]; the amount of blinking [
16]; the frequency and duration of a gaze [
17,
18,
19]; and the activity of the brain areas [
20], specifically the amygdala, which is involved, among other things, in the interpretation of facial expressions and emotions [
21].
Eye contact not only functions to provide information but also influences other domains. Research has shown that meeting a direct gaze affects cognition, for example, mental control, because some cognitive functions, such as mental control, are seen as irrelevant during eye contact [
22]. Additionally, a direct gaze has been shown to improve several social cognitive functions, such as face memory [
23], joint attention [
24] and empathy [
25]. Furthermore, eye contact results in affective reactions on the part of the receiver, such as feelings of intimacy, threat and arousal [
20,
26,
27,
28]. It also causes judgments of others depending on eye behavior, for instance, lower or higher ratings of potency, intelligence, attractiveness and credibility [
29,
30,
31,
32]. Due to the effects that eye contact has on the interaction partners, it also partly determines the content of their relation [
33]. As Laing formulates, “[w]hen we make eye contact, we experience a form of interpersonal connection that plays a central role in human social life, communication, and interpersonal understanding [
6].
The above shows that the way eye contact is established and how it is interpreted affects both the interaction partners and their mutual relationship. Decades of research have shown that a distinction can be made between normal and atypical eye contact. Atypical eye contact is mainly described in people with psychiatric disorders, such as autism spectrum disorder [
34], social anxiety disorder [
35,
36], schizophrenia [
37,
38] and bipolar disorder [
39]. Atypical features are, for example, staring too long [
40], gaze avoidance [
41], impairments in recognizing emotional expressions [
42], impairments in gaze discrimination [
43] and negative feelings that are induced in the receiver because of eye contact [
44].
The question may arise of the basis on which these features can be seen as atypical since they also occur in people without psychiatric disorders. For example, major differences have been found between nationalities in how eye contact is presented by the sender and interpreted by the receiver. Studies using eye-tracking methods have shown that East Asians look toward the center of a face, while Westerners alternate their focus between the points of a triangle, formed by both eyes and the mouth, when learning and recognizing faces [
45,
46]. Australians have a longer duration of eye contact than Japanese [
47]. This difference may correspond to the observation that maintaining eye contact is appraised positively by Western Europeans but not by people with an East Asian cultural background [
48]. Consistent with this, previous studies have shown that Japanese show less eye contact than Canadians during face-to-face interactions [
49,
50]. Compared to people from the Caucasus, Asian people have also been found to stare less at each other during social interactions, which has been attributed to the sociocultural norm of “gaze avoidance”, where an averted gaze functions as a sign of respect in Asian cultures [
48,
51]. In addition to culture, different features of eye contact have also been found in other social categorizations, such as in gender [
52], age [
53] and personality [
54,
55].
The above shows that a large amount of research has been conducted into how eye contact is presented and interpreted within various social categorizations and psychiatric disorders. For this scoping review, all these presentations can be understood as either the eye behavior of the sender during interactions or interpretations of eye behavior by the receiver. To gain insight into the overarching variations of presentations and interpretations of eye contact in all human populations worldwide, it is necessary to collect and combine all available research articles that focus on both social categorizations and psychiatric disorders. A preliminary search is performed for existing systematic reviews and scoping reviews on mapping the total forms of presentations of eye contact in adults per social categorization and psychiatric disorder as well as the effects of eye contact on the interaction partner. Searches were performed in the following databases: JBI Evidence Synthesis, Cochrane Database of Systematic Reviews, PubMed and AnthroSource. This process showed that there is currently no review available on this topic. For this study, the scoping review methodology developed by Peters et al. [
56] was chosen because of its suitability to locate, review and synthesize the existing literature on the topic through systematic searches [
57]. Using inclusion criteria based on the PCC elements (population, concept and context), the following question was used: “What is known from the existing research literature about variations in appearances and interpretations of interpersonal eye contact in social categorizations and psychiatric disorders worldwide?” This question was subsequently divided into two sub-questions: (1) “What is known from the existing research literature about variations in appearances and interpretations of interpersonal eye contact in social categorizations worldwide?” and (2) “What is known from the existing research literature about variations in appearances and interpretations of interpersonal eye contact in persons with psychiatric disorders worldwide?” The aim of this review is to synthesize the relevant literature, which can lead to a better understanding of the variations in the usage and meaning of eye contact worldwide and provide information for future research.
2. Method
This research aims to provide a broad overview of how eye contact is presented and interpreted in adults worldwide, both in social categorizations and psychiatric disorders, and to assess the quality of the included articles. The study was conducted according to the Joanna Briggs Institute (JBI) methodology for scoping reviews [
56,
58] and the provided checklists [
59]. For the article, the Preferred Reporting Items for Systematic Reviews and the Meta-Analyses Extension for Scoping Reviews (PRISMA-Scr) guidelines [
60] were used.
Prior to the study, the inclusion criteria, search strategy, handling of screening, extraction and synthesis were recorded in an a priori research protocol, which was created according to the PRESS 2015 Guideline Statement [
61] and published prior to the review in the Open Science Framework (ID: osf.io/7htsx).
Concerning eligibility criteria, the selected studies for research sub-question 1 included participants in adulthood (18+) from every social categorization worldwide and one of the following three conditions: (1) variations in appearances of interpersonal eye contact in social categorizations, (2) interpretation of variations in appearances of interpersonal eye contact by others or (3) correlations of appearances of interpersonal eye contact and its interpretation by others. The selected studies for research sub-question 2 included participants in adulthood (18+) diagnosed with psychiatric disorders, according to an internationally recognized classification system, and any one of the following three conditions: (1) variations in appearances of interpersonal eye contact in psychiatric disorders, (2) interpretation of variations in appearances of interpersonal eye contact by others or (3) correlations of appearances of interpersonal eye contact and its interpretation by others. For both research sub-questions, an open context, English and Dutch language, and no date limits were applied.
As described in the research protocol, relevant keywords, medical subject headings and search strings were identified for research sub-questions 1 and 2 during search trials in the databases MEDLINE and CINAHL and a peer review assessment by a medical-information officer (TI). Considering that the scoping review was not a comparative study, the search strategy was based on the building-block approach (BBA). See
Supplementary Materials,
Tables S1 and S2 for an example of the full search strings.
Databases were searched from inception until 25 May 2023. For research sub-question 1, the included databases were AnthroSource, Medline, CINAHL and the Psychology and Behavioral Sciences Collection (EBSCO). For research sub-question 2, the included databases were Medline, CINAHL, the Psychology and Behavioral Sciences Collection and PsycInfo. For searches in the grey literature, the first 10 pages on Google Scholar were used for research sub-questions 1 and 2.
Included characteristics of the sources of evidence for both research sub-questions 1 and 2 were randomized controlled trials (RCTs), non-randomized controlled trials (non-RCTs), quasi-experimental before and after studies, prospective and retrospective cohort studies, case-control studies, analytical cross-sectional studies, qualitative studies and reviews and research articles in the grey literature. In each article, only sub-studies that met the inclusion criteria were included.
For handling and summarizing data variables, a data-extraction form was used after being tested by the research group on 25 articles for usability and as a measurement of assumptions. After approval, two researchers (J.B. and R.S.) independently screened titles and abstracts then full texts.
Data for research sub-questions 1 and 2 were extracted on the following article characteristics: author; year of publication; title; affiliation of the first author; location of study; journal; medium; full reference; type of evidence; study design; sub-study (of each included study; only sub-studies that met the inclusion criteria were included, and if there were multiple outcomes or populations within that sub-study, these outcomes were divided); intervention; measurement instruments; aim of article; number of participants; size of the intervention group; size of the control group; presentation or interpretation; population; age and gender. When data were available, mean age was primarily noted, then range, and when no details were available, the field was rated with “adults.” When characteristics were not available, fields were rated with “unspecified.” If the source was not suitable for specific data, it was rated “not applicable.” See
Supplementary Materials,
Tables S3 and S4 for templates of the data-extraction form for research sub-questions 1 and 2.
The systematic assessment of evidence by means of a critical appraisal is not mandatory within scoping reviews [
62]. However, because of the added value of assessing the trustworthiness, reliability and bias of evidence, a critical appraisal was conducted (by J.B.), according to the JBI criteria [
59]. Articles were included if they met more than half of the JBI criteria.
The data on article characteristics were extracted and thematically analyzed. From this process, variations of presentations and interpretations of eye contact in social categorizations were distinguished and clustered for research sub-question 1. Simultaneously, variations of presentations and interpretations of eye contact in psychiatric disorders were identified and clustered for research sub-question 2 (psychiatric traits were assigned to research sub-question 1 due to the lack of an inclusion criterion for official diagnoses). The software programs Endnote version 21.2 and Microsoft Excel version 3208 were used during this process. The article was written using Microsoft Word version 2308 and Zotero 6.
4. Discussion
This review synthesized the relevant literature on variations in the usage and meaning of eye contact in social categorizations and psychiatric disorders worldwide and provided information for future research. Critical appraisal according to JBI criteria was applied to determine the quality of the included studies in this review. For both social categorizations and psychiatric disorders, the included studies met the majority of the JBI criteria. The lack of meeting all criteria was mainly in studies with quasi-experimental and RCT designs.
A novelty of this scoping review is that it provides an unprecedented broad overview of the existing knowledge about variations in appearances and interpretations of interpersonal eye contact in all possible populations worldwide. The results show that there are many variations in social categorizations and psychiatric disorders worldwide, with the vast majority of outcomes focusing on populations without psychiatric disorders. It remains unclear why populations use specific forms of eye contact or respond to them in a specific way and for what reason populations do or do not share variations in appearances or interpretations.
One possible reason is that existing research does not include sufficient factors when examining dyadic eye contact. Results from this scoping review show that variations are influenced by multiple characteristics, such as nationality, age, personality, sex and psychiatric traits. This finding indicates that effective insight into eye contact in populations requires that all influencing characteristics of these populations and between populations should be weighed. However, almost every study focuses on a singular characteristic. Additionally, there are some characteristics that may influence eye contact but which have received little attention in studies thus far. For example, of the total of 386 articles, only one study focuses on an African country [
152] and one study on a South American country [
49], which raises the question of to what extent eye contact is a point of attention in these continents. No research was found with a focus on religion, and very limited research was identified with a focus on cultures, despite previous research having shown that approaches to eye contact partly arise also from cultural contexts [
89,
90,
91]. As with social categorizations, no final insight has yet been obtained into the cause, effect and management of eye contact in psychiatric disorders. A reason for this may be that research to date has mainly focused on autism, schizophrenia and social anxiety disorder, while aberrances have been described in many other disorders. Another reason could be that the views and experiences of participants regarding eye contact have not been inventoried, with the exception of one study by Trevisan et al. [
44].
Because the studies found do not provide a clear image of how eye contact should be presented and interpreted worldwide, this could raise the question of whether there is such a thing as a “gold standard” for how eye contact should be managed. Although similarities are found within nationalities and characteristics such as gender in how eye contact is conducted, the totality of outcomes in this review suggests that a global gold standard does not exist. Rather, managing eye contact seems to arise from an individual combination of influencing characteristics that determine social meaning, which is consistent with previous research, such as that of Hamilton [
110].
The fact that no gold standard has been found to date also makes it difficult to determine on what specific basis appearances and interpretations of eye contact must be seen as atypical in populations with, compared to without, psychiatric disorders. Following the previous observation that approaches to eye contact arise from social meaning, eye contact in psychiatric disorders may better be seen as a spectral deviation from social norms rather than a symptom of illness. From this perspective, people who experience inconvenience because of eye contact would benefit from future research that focuses more on the consequences of eye contact on their perceived quality of life.
4.1. Strengths and Limitations
One strength of this scoping review is that it provides, for the first time, a transparent and reproducible overview of existing research on presentations and interpretations of eye contact in all possible human populations worldwide. Additionally, the current scoping review shows that there is a large amount of variation in presentations and interpretations, not only between populations but also within populations. Another strength is that potential bias is reduced by the critical appraisal of study quality, the use of peer review during the selection process and the calculation of inter-rater reliability.
There are also some limitations that must be considered. For example, this scoping review assumes dyadic eye contact in the broadest sense of the term. However, the debate is still ongoing about what the precise definition of dyadic eye contact should be [
4,
5,
6,
153,
154], so there may be differences of opinion about which studies should have been included. Another limitation is that although this scoping review analyzed many articles, important articles on presentations and interpretations of eye contact may have been missed. This could be due to the different views on and descriptions of eye contact in articles, because of which, search terms were not met. Articles were also excluded if the participants (partly) consisted of children, but these articles may have yielded meaningful data. Additionally, articles may have been missed because only Dutch and English articles were included, which could also explain why almost no studies were found from the African and South American continents in both social categorizations and psychiatric disorders. Another limitation is that in much research, only limited descriptions were given of the social categorizations of participants. Therefore, a less detailed oversight may have been given than desired. For example, out of necessity, the country in which the research was conducted was noted when the nationalities of participants were not clearly described.
4.2. Conclusions
An important conclusion of this scoping review is that there are many variations worldwide in how eye contact is presented and interpreted. This review identified 386 articles that contained 603 different outcomes on a wide range of populations, subdivided into social categorizations and psychiatric disorders. Overall, methodological quality was moderate to good. The prior literature has mainly examined presentations or interpretations of eye contact in specific populations. The diversity of these interpretations has been confirmed in this scoping review, but the current study also shows that there is significant variation between and within populations. Another conclusion is that aberrances of presentations and interpretations of eye contact are found in populations with and without psychiatric disorders. These findings also suggest that eye contact in psychiatric disorders could be a transdiagnostic factor that allows for a wide range of variation. Further, a gold standard for how eye contact should be presented and interpreted does not clearly emerge from the studies, so it is questionable to make general statements about abnormalities in eye contact. A shortcoming seems to be that virtually no research has been found that gives reasons for variations in presentations and interpretations of eye contact, how these variations are exactly linked to a specific population, or why multiple populations have these variations in common. One reason for this lack could be that, to date, very limited research has been conducted in which the views or experiences of participants regarding eye contact are inventoried. These described gaps could provide inspiration for future research.