Introduction
In 2019, a new infectious disease caused by the SARS-CoV-2 began spreading across the world. The World Health Organization (WHO) named it the coronavirus disease (COVID-19). Against the backdrop of the COVID-19 pandemic, wearing a mask has become a common phenomenon, since it is an important tool to effectively block the spread of the droplets that cause this disease. However, wearing a mask may have negative effects as well. Some studies recruited normal adults as participants and have found that masks provide a somewhat obscuring effect on the face, as they cover the nose and mouth areas of the wearer’s face, which is believed to affect the recognition of facial emotions (Carbon, 2020; Noyes et al., 2021) and influences one’s judgment of the wearer’s trustworthiness (Cartaud et al., 2020).
Cone of direct gaze is an index of the feeling of being looked at
The eye region, a facial area that is not covered by masks, shows the direction of eye gaze and can provide important information such as attentional location, behavioral intention, and emotional state (Emery, 2000; Mareschal et al., 2013b; 2014). Perceiving that a person is looking at us (Bateson et al., 2006; Carbon and Hesslinger, 2011; Hamilton, 2016), or looking away from us (Hietanen et al., 2008; Colombatto et al., 2020), has different effects on our behavior and perception. Therefore, an accurate perception of the gaze direction of one’s eyes is crucial to social interaction. However, previous studies have found that the perception of one’s gaze direction may not be very accurate. Humans prefer to judge others by looking at them (Mareschal et al., 2013a). To measure this tendency, Jun et al. (2013) conducted a study where participants were instructed to judge whether they perceived the gaze on faces looking in various directions to be looking at them. This index is commonly known as the cone of direct gaze (CoDG) (Gamer and Hecht, 2007; Jun et al., 2013). The wider the CoDG, the more the participant perceived the face as “looking at me.”
Factors influencing cone of direct gaze
Although CoDG is a relatively stable indicator for individuals (Lobmaier et al., 2021), it is influenced by a variety of factors. Firstly, facial information, such as head orientation (Gamer and Hecht, 2007), facial attractiveness (Kloth et al., 2011), facial expressions (Ewbank et al., 2009; Gillian et al., 2012), and so on, can influence CoDG. Secondly, CoDG is also affected by individual differences among the participants. People with autistic traits are less likely to judge being looked at, which means their CoDG is narrower (Matsuyoshi et al., 2014). Anxious individuals (Harbort et al., 2013) and schizophrenic patients (Wastler and Lenzenweger, 2018) are more inclined to judge someone looking at them as compared to others, meaning that their CoDG is wider. In addition, a recent study found that emotional situations may also affect CoDG (Rimmele and Lobmaier, 2012; Lyyra et al., 2017; Syrjämäki et al., 2017). Specifically, external pressure affects perceptual judgments of being looked at; being socially excluded increases one’s judgment of being looked at if there is a possibility of re-interaction (Lyyra et al., 2017) while it deceases if there is no interaction (Syrjämäki et al., 2017).
The current study
To sum up, CoDG is affected by emotional situations and wearing a mask can create a safe or a threatening situation, especially in the context of the COVID-19 outbreak. Thus, a CoDG task was conducted to examine how wearing a mask impacts the judgment of gaze direction. Different types of masks create different levels of perception of danger or safety (Chu et al., 2020). For example, N95 masks provide a higher level of protection from viruses than regular surgical masks. Unlike a surgical mask, an N95 one may be more likely to convey the message that the virus will not be spread by the individual wearing it. At the same time, wearing N95 masks may also mean that the surrounding environment is dangerous and the likelihood of infection is high. Therefore, three types of masked faces (non-masked faces, surgical-masked faces, and N95-masked faces) were considered in our study to further explore whether the effect of masks on CoDG is modulated by the type of mask being used. In addition, to investigate whether the impact of masks on CoDG was related to COVID-19 or a person’s individual characteristics, we asked the participants to complete a three-item questionnaire on COVID-19, Social Interaction Anxiety Scale (SIAS) (Mattick and Clarke, 1998), and Self-Rating Depression Scale (SDS) (Zung et al., 1965), after the CoDG task.
Given that previous research has already shown that wearing a mask can reduce social distance and increase willingness to socialize in the context of the COVID-19 pandemic (Cartaud et al., 2020), we predicted that participants will over-report their sensation of being looked at. Namely, the CoDG of the masked faces will be wider than that of the unmasked faces. In addition, previous studies have found that while trustworthiness or threat can widen the CoDG, wearing a mask may enhance the wearer’s sense of trustworthiness or threat. Thus, we predict that N95-masked faces may produce a wider CoDG than a face wearing a regular surgical mask.
General discussion
We conducted a gaze discrimination study to investigate the influence of wearing masks on CoDG. The results showed that wearing masks increases the width of CoDG. We also found that the CoDG of masked faces was significantly wider than that of unmasked faces. This indicated that wearing a mask would affect the processing of gaze direction. However, mouth-cut faces did not produce the same widening effect on CoDG as the masked faces. These findings show that the influence of masks on CoDG can mainly be attributed to the high-level social significance of wearing a mask, rather than the low-level physical information of the face.
Why does wearing a mask influence the judgment of gaze direction? One possible explanation is that wearing face masks increases the sense of trust and, thus, widens the CoDG. During the COVID-19 pandemic, most governments as well as the WHO recommended wearing face masks as a key measure to protect people from the novel coronavirus (Bhardwaj and Agrawal, 2020; Carbon, 2020). In addition, studies have found that, in the context of COVID-19, a face covered by a mask may be considered safer and more trustworthy (Cartaud et al., 2020). The increase in trust makes people willing to interact and, thus, more likely to judge that a face wearing a mask is looking at them. On the contrary, another explanation is that wearing face masks increases the sense of threat and, thus, widens the CoDG. During the COVID-19 pandemic, wearing face masks implies a risk of COVID-19 transmission for the wearer probably being a COVID-19 patient. Consequently, the mask could be perceived as a threat message highly relevant to one’s health. When facing threatening facial expressions or in a threatening situation, previous studies have shown that people tend to judge others as looking at them, resulting in a wider CoDG (Ewbank et al., 2009; Tso et al., 2012; Jun et al., 2013; Harbort et al., 2017). Previous studies have found that masks both increase trust and shorten the social distance, but also increase perceptions of sickness, possibly because the internalized rule of wearing a mask inhibits automatic evaluation of mistrust (Rosa, 2020).
The result of these two contradictions may be that people themselves may hold ambivalent attitudes toward masks. From a cultural psychological perspective, in the context of the COVID-19 pandemic, masks are given some social significance and are no longer neutral objects. Meanwhile, the mask becomes part of the body by covering the nose and mouth areas. It is an interface that simultaneously distances and connects me to the other. By wearing a mask, the person generates different levels of meaning and automatic hetero-regulatory processes (Tateo, 2021). Therefore, we did not specifically discuss whether the widening effect is due to an increased sense of security or threat from masks during the COVID-19 pandemic.
Besides, no significant impact of the mask types on the CoDG emerged; as the same widening effect on the CoDG was observed for both the N95-masked faces and surgical-masked faces. On the one hand, during the pandemic, people have been highly sensitive to the threatening information about COVID-19. As for information about the level of threat, masks that offer lower protection (surgical-masked) and those that offer higher protection (N95-masked) can induce the alert response from individuals. Therefore, the N95-masked and surgical-masked faces have the same widening effect. Apparently, such an undifferentiated alert response has adaptive significance for human survival. To give an example, although there was a difference in the feasibility of transmitting the virus between COVID-19 patients and those who had recovered, there was no difference in viewing time to their faces (Federico et al., 2021). In general, individuals have the same avoidance response to novel coronavirus-related information at different threat levels.
On the other hand, there is no significant difference in the cognition of mask protection in the individual’s perception. This may have something to do with advice circulated by the media to the general public to wear face masks during the pandemic, that both surgical masks and N95 masks provide adequate protection against novel coronavirus transmission through droplets (Sureka et al., 2020). Furthermore, people are equally familiar with and are likely to be exposed to both types of masks. Consequently, this may result in the inability to observe the different effects of mask types in our study. Future studies may use more diverse mask types having different protective characteristics (such as gauze masks vs. N95 masks) to further explore this issue.
The present study did not find any correlation between the three-item questionnaire on COVID-19 and scale and the mask effect, which was consistent with the results of previous studies (Lobmaier et al., 2021). The correlation results might be related to the sample size, and future studies could use larger samples to verify the correlation between individual traits and the mask effect. Another possible reason was that our participants were selected from college students and did not include clinically diagnosed patients, so masking the correlation between the questionnaire and CoDG. This could be further explored in the future by selecting clinical participants.
Another key point to remember is that there were some limitations in the participant selection process, since all of them were recruited from college. Inevitably, age and knowledge of background may have influenced their gaze judgments. Considering this, it would be useful for future studies to recruit people of different ages and backgrounds. It must also be noted that the participants were all Chinese. Previous studies have reported the significance of the culture of face masks (Timpka and Nyce, 2021) and different cultural effects on the perception of facial information (Blais et al., 2008; Jack et al., 2009). Meanwhile, during the COVID-pandemic, different cultures took different measures and also showed different attitudes to wearing masks. As such, further studies are required to compare the effects of face masks on gaze perceptions in different cultures. In addition, considering that taking physical masking approaches would introduce new variables (e.g., color, personal preference, etc.), the current study created mouth-cut faces by editing out the lower part of the models’ faces as the control condition. However, this operation could disrupt the integrity of a face. Previous studies found the presentation of a whole face affects the processing of face information (Leder and Carbon, 2005). Thus, further studies could seek out a better control condition to explore the widening effect of face masked on CoDG.
Furthermore, with the global outbreak of COVID-19 pandemic and the virus mutating, people have become fully aware of the seriousness of the epidemic and the wearing of masks has become widely recognized during the period of data collection (March 2021 to December 2021). Many countries have now announced the removal of epidemic prevention and control measures, and it is possible that people’s perceptions of masks may vary, so the results may not apply to those who are not required to wear them.