embarrassment as a medium of social interaction
and behavioural displays of impression management
Ph.D. Candidate in Media and Arts Technology programme,
School of Electronic Engineering Computer Science
Queen Mary University of London
"A self-idea of this sort seems to have three principal elements: the imagination of our appearance to the other person; the imagination of his [sic] judgement of that appearance, and some sort of self-feeling, such as pride or mortification." (Cooley, 1922: 184)
This research proposal suggests to study the relationship between a self-concept idea of Cooley ( 1922), Looking-glass self (LGS), and embarrassment, which come from a minor bodily stigma (e.g., facial scars) in a speed-dating scenario. The main objectives of this research are to investigate: 1) how embarrassment plays a critical role in human life to maintain social interactions and communications flow; 2) how a stigmatised individual’s embarrassment would be reflected and manipulated by an encountered individual’s (i.e., an observer) reactions. In order answer to these questions, this research will focus on finding behavioural evidence of this relationship. Specifically, this research will investigate the relationship between self-touching (by the stigmatised individual) and gaze direction (by the observer), which are well-known behavioural displays of embarrassment; and how this relationship affects both individuals' impression evaluations.
Looking-glass self (LGS) is a self-concept idea formulated by Cooley ( 1922) which argues that an individual's self-awareness involves constant self-evaluations of how others perceive them. Cooley described the process behind this concept as having three principal elements, which are described in the above quotation. The first two elements of LGS have often influenced the study of the processes of human representation and social interaction, which are using affective and behavioural measures: Interpersonal distance (e.g., Bailenson, Blascovich et al., 2001); eye gaze synchrony (e.g., Bailenson and Blascovich, 2008; Bailenson, Beall et al., 2005); nonverbal mimicry (e.g., Bailenson and Yee, 2005); and prosocial behaviour and verbal mimicry (e.g., van Barren et al., 2003). However, in relation to self-behavioural change, the third element, self-feeling has rarely been investigated in previous social psychological studies.
One of the self-feelings resulting from the LGS process, which Cooley specifically mentioned, was mortification; often referred as shame or embarrassment in modern studies. Goffman (1959) proposed that embarrassment over shame management is the central concept of Cooley's idea. Goffman argued that all social interactions in human life involve embarrassment; people are always managing their behaviours in order to avoid embarrassing experiences (i.e., self-impression management). Embarrassment occurs when people unintentionally violate social norms (or rules) in everyday human life. Goffman also insisted there is a strong connection between stigma (e.g., physical deformities; racial characteristics; political orientations) and embarrassment, which people could be involved by both voluntary and involuntary activities (Goffman, 1963; Ellis, 1998). Although Goffman mentioned a few minor bodily stigmas, such as scars or misshaped noses, the related studies have rarely explored this issue (Ellis, 1998).
Previous researchers have shown people express their embarrassment through certain behavioural signs such as blushing, aversion of eye contact, nervous smiling or laughter (e.g., see Edelman and Hampson, 1979, 1981; Goffman, 1956). Goffman (1956) argued a display of embarrassment could help the embarrassed individual's impression positively. Goffman noted that a display of embarrassment "… demonstrates that, while he (the actor) cannot present a sustainable and coherent self on this occasion, he is at least disturbed by the fact and may prove worthy at another time. To this extent, embarrassment is not an irrational impulse breaking through socially proscribed behaviour but part of this orderly behaviour itself" (p. 270). As such, not only the embarrassment displays but also the meaning of these expressions have been investigated in the field.
However, very few studies have been investigated the meaning of behavioural interactions between individual and another person (e,g., an observer) (e.g., see Miler 1987). Not only the individual's displays of his or her embarrassment but also the others' behavioural signs would influence the individual to experience embarrassment. Goffman (1956) insisted that embarrassment is not always experienced by an individual who experiences a difficulty; it could be for pairs who are together having difficulties; in some other cases, the observers could likely feel embarrassed for the individual. Therefore, it is important to understand how behavioural interactions between the embarrassed individual and encountered others occur.
A Speed-dating scenario seems a good social environment to study displays of embarrassment and behavioural interactions between two individuals; and how a pair who have not met previously try to convey favourable impressions to each other. Therefore, it would be interesting to examine how an individual constantly manages his or her embarrassment of stigma in this social interaction, also how an observer's behavioural displays would be expressed. Eibl-Eibesfeldt (1989) discovered some common behavioural connections between flirting and embarrassment displays (e.g., coy glances and smiles, face-touching, neck presentations). However, these favourable displays resulted from embarrassment, which occurs during socialisation situations and not from stigma. Hence, this research would also provide a good opportunity to investigate causal relationships between these behavioural expressions.
This proposed research will allow people to have a better understanding of social interactions and communications process in their everyday life. Furthermore, this investigation will be increasingly important to develop emerging technologies that mediate social interaction. For example, Ranganth et al. (2013) investigated human speech patterns (e.g., friendly, flirtatious, awkward, and assertive speech) using speed-dating scenarios. The research aimed to develop the computational natural language of speech processing and social computing by studying human’s social interaction. Bailenson and Blascovich (2008) explored the LGS by using immersive virtual environment (IVE) technology, which has shown new opportunities for manipulating appearance created by the IVE. Also, a collaborative company with the Microsoft HoloLens studio, Object Theory (2015) has launched their service to provide mixed reality (e.g., virtual reality or augment reality) experiences. With a shared holographic experience using by computational devices (e.g., PC, tablet or mobile), multiple users could interact with each other regardless of location. More recently, Microsoft Research (2016) presented Holoportation project, which suggested a live holographic social interaction experience through the HoloLens system. As such, this proposed research would expand new opportunities to such technologies to evolve in the future.
Figure 1. A diagram of the observer’s gaze direction display reflected in the self’s embarrassment display (face-touching)
In this proposed research, I will focus on individuals who have a scar on their face. The research will investigate how this minor stigma influences their social interactions with others (see Figure 1). Therefore, the key research questions are stated as below:
Does the breaking of eye contact by the observer initiate self-touching by the individual?
Does the position of the observer’s gaze dictate the position of the individual’s self-touching (e.g., face touch or hair touch)?
3. Do these behavioural expressions (self-touching frequency; eye contact frequency) lead to favourable impressions (e.g., a proposal of the second date) in a speed- dating scenario?
Looking-glass self and embarrassment
Figure 2. An illustration of the Looking-Glass Self-concept (Rcragun, 2009)
Looking-glass self (LGS) was first formulated by Cooley in 1902 (Cooley,  1922). One of the fundamental arguments of this self-concept idea is that individuals see themselves through how others interact with them as a looking glass (Figure 2). Cooley described this as people "… live in the mind of others without knowing it" (p.208). Another proposal of this idea is that as a result of this imaginatively "living in the mind of others" process, individuals develop two real emotions, either pride or shame. Scheff (2000) explained that these two shared emotions are the most intense and powerful social emotions because they are deeply related to social bonds. These two emotions are socially reflective. Individuals experience the feeling of pride when the social bond is entirely undamaged, and the shame comes because of this bond is threatened or cut-off. However, pride and shame act differently. Pride is directly expressive, whereas shame is generally hidden, therefore it can often be unacknowledged or bypassed (Scheff, 2005b).
Goffman (1959) proposed embarrassment as another social emotion that results from the LGS process: the self-impression management of embarrassment. Goffman insisted embarrassment plays a communication role in social interactions to help verbal and non-verbal cues to be successfully attuned. Goffman argued individuals have two selves in everyday life: the virtual self is a specific identity or self, which is formed by idealised expectations; the real self is formulating from the individual's behaviours. Scheff (2005a) referred to this role of embarrassment as "an interactional device", which makes individuals behave against their idealised expectations in everyday life. Scheff also defined the difference between embarrassment and shame as: "Embarrassment works among people, shame works within people" (Scheff, 2005a: 2). In this proposed research, I shall look into embarrassment resulting from shame that comes from a minor bodily stigma, to explore how this social emotion allows people to maintain social communications and interactions.
Stigma and embarrassment
Goffman (1963) argued that individuals carry a stigma when they have traits, which persistently damage their identity and stop full-participation of social interactions. Goffman described that a stigma is spoiling an individual's identity.
There are three types of stigma that Goffman presented: 1) imperfections of bodily systems; 2) defects of individual's character perceived as a lack of will or passions (e.g., addictions or mental disorders); 3) memberships of a minority group (e.g., race; nations; religion). These stigmas would bring negative impressions, therefore the stigmatised individuals always try to prevent themselves from having to deal with the normals (i.e., “particular expectations at issue”). These stigmas could be concealed by individuals. For example, limiting social activities or involvements could be one of the ways, if they could be hidden by this strategy. However, certain stigmas could not be concealed because they are visible. In this case, the stigmatised individuals attempt to control their status indirectly by giving themselves more comfortable areas of ordinary activities. For instance, and generally speaking, a blind person would not try to be an expert in skiing or mountain climbing. Many existing studies have investigated the relationship between bodily-stigma and social emotion (embarrassment), however, they were mainly focused on populations with major bodily issues: Older deaf people (Becker, 1980); dwarves (Ablon, 1984); wheelchair users (Cahill and Eggleston, 1995).
However, many people have minor stigmas such as scars, eye problems, body odours, constant facial-blushing or voice tones (Ellis, 1998). These minor stigmas could also be associated with embarrassment. Hence, this research proposal suggests to investigate how individuals' embarrassment displays would be presented by a minor-bodily stigma (a facial scar); and what links would be discovered in relation with the LGS process.
Embarrassment is not an irrational impulse breaking through socially prescribed behaviour but part of this orderly behaviour itself. (Goffman, 1956, p. 271)
Some non-verbal displays of embarrassment have been recognised as common behaviours that people exhibit when reacting to an embarrassing event: Blushing; smiling; laughing; avoiding eye contact and self-touching (Edelmann, 1987; Harris, 2001).
Most previous studies investigated visible behavioural displays. Edelmann and Hampson (1981) reported that embarrassed individuals decreased their eye contact and presented more gestural activities and smiling. Keltner's study (1995) also showed that participants have shown some particular behaviours such as embarrassment displays: Constant shifts of gaze, looking down, smiling control, or touching face. A study by Asendorpf (1990) looked into the relationship between gaze and smiling while embarrassed; gaze aversion occurs before the apex of the smile (when the end of mouth as at its most extended).
There have also been investigations on physiological change during an embarrassment event, since Darwin ( 1998) insisted blushing is an indicative expression of embarrassment. Only a few studies have shown the correlatives between blushing behaviour and embarrassment - for instance, conducting cheek and ear colouration as well as temperature change through photoplethysmography (Shearn et al., 1990; 1992). However, these results have rarely been accepted from the field of psychology studies, as many limitations of measuring the change have been reported - for example, individuals could show different results by their skin or hair colours; children do not easily blush (Richardson, 2013, p.291).
Self-touching as an embarrassment display
Self-touching refers human body or face contact, which generally through the hand, includes scratching, rubbing, caressing, and grooming actions. These behaviours are not usually displayed to convey communicative intention; with little or without awareness (Harrigan, Kues and Weber, 1986). Existing studies have shown that self-touching displays when individuals experience negative affect, anxiety, discomfort, or conflict. These behaviours surprisingly perceived as positive that observers could feel more intimate to the individuals (Harrigan, Kues and Weber, 1986; Harrigan, Weber and Kues, 1986).
Although a few studies reported self-touching (e.g., face-touching or body touching) is one of the displays of embarrassment, it has rarely been investigated. In Keltner's (1995) study, face-touching was measured as a part of a components study approach - seven different nonverbal cues were measured (gaze down; gaze shifts; smile controls; smile; head movements; head turn to side; face touches). The study compared how these behavioural displays of individuals are expressed while they experience either embarrassment or amusement. The study showed that face-touching was more frequently occurred when participants experience embarrassment than amusement (26% vs. ll%,r= 1.47,/x.lO). However, the self-touching was the least expressive display among seven components. Keltner argued that this result seemed to be reduced by the physiological recording devices, which were attached to the individuals' hands.
The others' displays of embarrassment
Many existing studies have shown that observers could recognise the displays of embarrassment easily; they could discern these displays from other emotional expressions (e.g., shame) (Keltner 1995; Keltner and Buswell 1996, 1997). There seem few studies which investigate the observers' behaviours, which could be displayed as a sign of response to the actor's embarrassment. Miller (1987) presented two studies, which demonstrated how the observers who witness the individual's embarrassment responds both cognitively and physiologically (by recording skin potential). The reported results of both studies showed that the observers experienced empathically shared embarrassment with the actor. However, these empathic observers did not think that they felt the embarrassing actions of actor’s were directly reflected on them. Miller described that although the actor's behaviour indeed influenced the observers' perception of the embarrassment, the empathetical aspect is not completely dependent on the actor's embarrassment display. The observers judged the actor's embarrassment based on the social norms, which could allow empathic embarrassment to appear.
Purpose or Hypothesis
As seen from the reviewed literature, the majority of studies that are related to embarrassment displays have been focused on embarrassment which occurs during socialising situations. However, the embarrassment displays from bodily-stigma have rarely been investigated (Ellis, 1998). Since bodily-stigmatised individuals, who’s stigma is visible (even if it is minor), are already aware of their differences, the displays would appear differently from situational embarrassing actions. In this, the observers' behavioural displays may be a huge influence on the individuals' embarrassment. In this proposed research, I would like to focus on facial scarring as a minor stigma for the individual. I hypothesise that the individual who has a scar on their face would be touching their face or hair frequently when they are socialising. I would also like to investigate how the observers' gaze directions would be influenced by this individual's embarrassment display changes. This behavioural changes in individuals would also unconsciously manipulate the observers' gaze directions.
Method of Approach
The first study will be an investigation of the research question 1 and 2, which are stated in the previous section. The study will investigate: 1) the influence of the observer’s eye contact breaking on the stigmatised individual’s self-touching display frequency; 2) the influence of the observer’s gaze direction on the individual’s self-touching position.
To investigate these research questions, I shall use immersive virtual environment (IVE) technology. The method will allow participants to interact through virtual human representations: Virtual representations of self (VRS); virtual representations of others (VRO) (Bailenson and Blascovich, 2008). IVE technology has been used in many social psychology studies because the method allows researchers to manipulate the virtual representations accurately (Bailenson and Blascovich, 2008; Blascovich et al., 2002; Loomis et al., 1999). Existing studies have been supported the effectiveness of IVE technology to study human behaviours and various subjects of social interaction, including interpersonal distance (Bailenson, et al., 2001; Bailenson and Blascovich, 2008), eye gaze (Wieser et al., 2009; Bailenson et al., 2001, 2005), nonverbal mimicry (Bailenson and Yee, 2005), persuasion (Guadagno et al., 2007), and social inhibition (Hoyt et al., 2003). These studies showed that people could interact with VRO as if they were with human since the VRO could present realistic nonverbal human behaviours (e.g., blinking; lip movement; eye contact).
Figure 3. Example of gaze direction control of VRO in IVE through the Poser (Wieser et al., 2009)
Therefore, I shall use a 3D software, Poser (smithmicro.com, 2016), which is a suitable tool to create 3D avatars, in order to design a VRO. The software has been used in a few social psychological studies, which dealt with gaze direction manipulations (e.g., see Wieser et al., 2009) (Figure 3). I will use an additional 3D software, Unity (Unity3d.com, 2015), the VRO to interact with a participant (i.e., stigmatised individual) through IVE.
During the study phases, the participant’s eye movement will be monitored by a remote eye-tracking system (e.g., [Senso Motoric Instruments, RED250; RED500; iView X Hi-Speed] SensoMotoric, 2016). An infrared light source will reflect the corneal, therefore it will allow for the measurement of the eye movements.
Since the study will investigate a facial scar as a minor stigma of embarrassment, the self-touching would appear around the face. Therefore, in this study, participant’s face-touching will be monitored. The face-touching display will be recorded by a webcam with an affordable 3D sensor (e.g., the Microsoft Kinect). An automatic hand detecting system will be used to analyse the 3D data (e.g., Mahmoud et al., 2011). In order to process this automatic detection through labelling the face-touching hand’s location, Elan video annotation tool will be used (see Lausberg and Sloetjes, 2009).
The main study will focus on the research question 3, based on the results from the study 1. The study will investigate the influence of the embarrassment displays (face-touching frequency and eye contact frequency) of a stigmatised individual and an observer on favourable impression management in a speed-dating scenario. The study will use a set-up environment of speed-dating, where a stigmatised individual and an observer could interact.
During the study, the gaze direction and hand location of both the individual and the observer will be monitored by the remote eye-tracking system and automatic hand detection system, which will be the same systems that were mentioned in study 1 section. A short question will be asked to the pair whether they would ask the second date or not in order to understand the relationship between these behavioural expressions and favourable impression.
This research will present critical data and analysis to understand the relationship between the LGS and embarrassment displays. The outcome will follow the aim of this study that explores the manipulation of a stigmatised individual’s embarrassment display (face-touching) frequency by an observer’s embarrassment display (eye contact breaking) frequency. The research will use advanced technologies to investigate these behavioural displays’ interaction including IVE technology, eye-tracking system and automatic hand detecting system. By using both IVE technology and real-event (speed-dating) based studies, this research will provide detailed measurements of behavioural expression and their interaction. This approach of outcome will contribute to the field of social psychology and cognitive science, which are investigating Cooley’s LGS concept and its association with self-feeling (embarrassment) displays. Moreover, the research outcome will provide new opportunities for developing various technologies that mediate social interaction and communication.
List of journals and conferences that would be suitable to present this research as below:
Journal of Applied Social Psychology
Cognition & Emotion
Journal of Experimental Social Psychology
Journal of Personality and Social Psychology
CHI (Conference on Human Factors in Computing Systems)
DIS (Designing Interactive Systems)
CogSci (Cognitive Science Society)
Computation, Communication, Aesthetics and X (xCoAx)
Tangible, Embedded and Embodied Interaction (TEI)
Inter-Society for the Electronic Arts (ISEA) Work Plan
...(download the rest of the essay above)