Abstract

    Throughout our everyday lives there are infinite times we interact with other people. Within these interactions, what are the cues that others use to make judgments about us? Research suggests that in many situations, such as job interviews, blind dates, and even in everyday general conversation, eye contact can be a huge factor in the judgment of another. Even in such circumstances as compliance to a request, the result can be affected by how much eye contact the person who is asking maintains. Given the importance of eye contact in our social situations, this study proposes there are certain other factors that are related to eye contact, and eye contact is a representation of these other factors. A relationship between eye contact and traits of personality and self-esteem would present a way to infer the levels of such traits by the amount of eye contact a person assumes in a conversation. For instance, a teacher could observe the amount of eye contact a student has in a meeting, and from this information could shed a small amount of light on the personality of a student.
    For this study, eight male and eight female participants were used from a pool of introductory psychology students at a small New England college. Participants were administered the Rosenberg Self-Esteem Scale (1986) and the NEO Five Factor Inventory (Costa & McCrae, 1992), a personality scale. Following the completion of these tasks, participants were fitted with the Eye Tracker (Sensomotoric Instruments), a head mounted device that tracks the movement and focus of the participant’s eyes and plots it on a video recording. Once the participant was calibrated to the Eye Tracker, the researcher left the room and a research assistant had a conversation with the participant, without the participant knowing the conversation was the main focus of the study.
    Pearson product-moment correlations were performed to find a relationship between the dependent variables. A significant relationship was found for self-esteem scores and eye contact (r = -.58, p = .017). Eye contact was measured as the number of times the participant broke eye contact during the conversation. There was no significant relationship between any of the five personality factors and eye contact, and an independent samples t-test yielded no significant differences between males and females. This data infers a negative relationship between self-esteem scores and the number of gaze breaks in a conversation. That is, as esteem scores increase, the number of breaks decreases, pointing to stronger eye contact.
Because of the small number of participants in this study, the results cannot be truly generalized to an entire population. However, future research could be done to find further support for the measure.

Back to top

Introduction

    Eye contact and gaze, members of a family of nonverbal communication cues, have been important factors in communication that have been iterated over time. For some people, simply establishing eye contact can be interpreted as an act of hostility or anger (Argyle & Cook, 1976). In some situations eye contact can also be interpreted in the opposite manner - as a sign of friendliness, romantic attraction or general interest (Argyle & Cook, 1976; Kellerman, Lewis, & Laird, 1989; Kleinke, 1986). Given this, one can say that eye contact is at least the grounds for minimal social interaction, and because of the widespread interpretations of simple eye contact, it would be an advantage to understand this nonverbal language. Actually, according to many researchers, the detection and interpretation of eye gaze plays a prominent role in both the development of social cognition and the smooth running of everyday social interaction (Baron-Cohen, 1994; Perrett & Emery, 1994).
As we progress into adulthood, eye contact begins to have more of an impact on our successes and failures in society. Similarly, esteem is a trait that can define the way we think of ourselves and also the way we think about others. High self-esteem brings us to believe that we are intelligent, attractive, and popular (Baumeister, Campbell, Krueger & Vohs, 2003). Levels of self-esteem even influence the way we assess another’s personality (Lagomarsino, 1998). An assessment of esteem by Crocker and Park (2004) examines the importance of self-esteem in today’s world. Within the article, the apparent importance contemporary society places on esteem is iterated. According to these authors, this trait has become a focal point of families and schools. Given the importance of both eye contact and esteem, is there a relationship that can be found between the two? The current study will assess this possibility.
    In all of psychology, self-esteem is perhaps the most widely studied personal dimension (Watson, Suls & Haig, 2002).  It is also considered an important and influential concept in personality (Bushman & Baumeister, 1998). Though there have been reports of relationships between esteem and personality traits, Watson, et al. insist that no true, empirical evidence of the relationship between these traits has been reported (Watson, et al., 2002). However there is some evidence that a relationship can be found between esteem and the lower tiers of the “big five” personality traits. That is, each trait can be broken down into more specific traits, and some relationships between these more specific traits and self-esteem have been established.  Watson tested this theory, and found that testing for the lower level traits allowed a more detailed interpretation of the results of a generalized self-esteem scale. For example, scoring low on a self-esteem inventory may not point to depression on its own, however coupled with the more detailed results from a neuroticism scale this conclusion could be sound. The current study will assess the relationship between esteem and personality, using the big five model of neuroticism, extraversion, openness, agreeableness and conscientiousness.
    The behavioral result of the blending of social and cognitive mechanisms is another reason to look into the meaning behind gaze. It has been shown to be useful to have an information processing system that can deal with perceptual input, such as reading facial signals and eye gaze, in a rapid manner (Baron-Cohen, 1994; Perrett & Emery, 1994). Not only do we need to recognize the presence of gaze in the environment in order to discern the intent of another individual, we also need to establish the identity of that individual - eye contact facilitates this. In a study by Guegen and Jacob (2002), it was demonstrated that eye contact could play a large role in the compliance of others to a request. Confederates approached subjects in a defined area, and asked them to answer questions about natural products. While making the request, the confederate used one of two techniques - they either maintained strong eye contact throughout the exchange, or averted their eyes as soon as the subject made eye contact. These tactics were repeated consistently throughout the exchange. This study found that a majority of people (66%) complied with the request to answer the survey when direct eye contact was used. Evasive glances only garnered 34% compliance. Interestingly, there was a gender difference in this study as well: with direct eye contact, 76% of women consented to the survey compared to 56% of the men, whereas in the evasive gaze group, the reverse was true; 24% of women complied to the request as opposed to 44% of the men. The gender difference in this study may be attributed to different things. Because a person is more likely to attribute a trait to another when their response to a situation differs from what he or she would do (Pronin, Gilovich & Ross, 2004), perhaps women refuse because they were not looked in the eye upon request. Previous biases to being treated “like a girl”, or without respect, could play a part in the response to this study. It would be interesting to note if the gender of the confederate had an effect on the response of the participant.
Within this study by Guegen and Jacob, the illustration of a relationship between eye contact and the compliance or rejection of a request is apparent. This implies a judgment on the part of the participant, who may be distinguishing a particular attribute in the person making the request. A relationship between eye contact and personality type, or even esteem, could lead to a reason for the compliance/rejection of a participant other than the assumed eye contact. For example, the participant who faced the confederate with little eye contact may have decided that the person was not amiable, or perhaps not sincere. Though these are just suggestions, they are judgments made about the confederate’s personality that could be made while actually interacting very little. Such judgments could be considered extremely inaccurate, considering the small amount of information garnered from such a short encounter; however, this raises the question, do people make such judgments? According to Lagomarsino, (1998) people do judge others, and can do so according to the level of eye contact they have.
    According to Lagomarsino, Gallagher, Yankalunas, Brooks & O‘Brien (1998), eye contact plays an extremely important role in the perceiving of a person's esteem, along with other factors. In this study, students were first asked to take the Multidimensional Self - Esteem Inventory (MSEI) to assess their self-esteem. Approximately four weeks later students were then asked to watch clips of film where a man was being interviewed. There were two versions of video, one in which only five seconds of eye contact was maintained and one in which eye contact was maintained for fifty seconds. Both tapes were sixty seconds long. Before watching these tapes, students were given a copy of the MSEI, and asked to fill out the form as they thought the individual in the video would have filled it out. The tape was shown twice to each participant. The results of this study were interesting. As eye contact of the person in the video increased, the projected self esteem for the model varied according to the self-esteem of the participant. For instance, participants in the high self-esteem group scored the extended eye contact model as having high self-esteem and the limited eye contact model as having low self-esteem. The opposite was true for those people who ranked themselves low in self-esteem; they rated the person with extended eye contact as low in self-esteem and the person with limited eye contact as high in self-esteem. However, there are alternate explanations for these results. Subjects could potentially be projecting their own level of self-esteem onto others, or possibly be projecting complementary traits. (Lagomarsino, et al. 1998). Because of the relative instability of the results from many studies such as this one, it is important to perform further research that investigates the relationship considered by Lagomarsino, et al.
    Along with instances such as the compliance study, there are also other social interactions where eye contact can be pivotal. In instances such as job interviews and blind dates, the impressions created during initial encounters govern one’s access to significant social and material awards - as well as subsequent interactions (Garcia, Stinson, Ickes, Bissonnette, & Briggs, 1990). We meet new people nearly every day, and our eye contact plays a role in the way people perceive us.  Brooks, Church, and Fraser (2001) looked at duration of eye contact by a character in a video, and the subsequent assessment of that character's personality by an observer. Brooks, et al. found that as eye contact increased in the video (from five seconds to thirty out of a total sixty second video) characters in the video were rated higher on personality traits such as assertiveness, decisiveness, dominance, and aggressiveness. Within this study, duration of eye contact plays a critical role in the appraisal of another person. Likewise, another study done in the job setting involved the appraisal of applicants based on their eye contact.
Some have gone so far as to say that confidence, control and positive emotional state are subject to levels of eye contact (Amalfitano & Kalt, 1997). In their study, Amalfitano and Kalt used two pictures of a male and female “job applicant”. These applicants had one picture where they were looking at the camera, and one picture looking down. Forty four random job interviewers from an employment agency rated the pictures on various traits, and also on whether or not they would hire the applicant. This study found that just from a picture, the raters found the confederates looking at the picture higher on certain traits such as confidence and control. The study also found that eye contact is significant in the hiring process. However, it remains to be shown whether the judgments made of such individuals are an accurate portrayal of their personality traits. The current study will test such judgments by assessing the relationship between the eye contact of a participant and their individual esteem and personality scores. In this way, the research can be furthered and such appraisals as those the job interviewers are making can be deemed accurate or inaccurate, which can prove beneficial to business owners and others.
    Through the studies explored previously, there is a lack of a definite connection between eye contact and such factors as gaze and personality, though the connection was hinted at. The lack of research on this particular area leads to the question presented by this study: what, if any, is the relationship between eye contact and the traits of personality and esteem? A direct relationship between self-esteem levels and eye contact, as well as a relationship between personality and eye contact are predicted. Also, a difference between gender in the area of eye contact is hypothesized.

Back to top

Method

        Participants
    The sample for this study was composed of 16 participants, 8 male and 8 female all between the ages of 18 and 22. The participants were a part of the subject pool for a credit in introductory psychology classes at a small liberal arts school in New England.
        Measures
    Participants completed the NEO - Five Factor Inventory, developed by Costa and McCrae (1992). This tool assesses the participant on a five factor model of personality, yielding five self reported scores. Internal consistency calculated for the NEO-FFI was .86, .77, .73, .68, and .81 for neuroticism, extraversion, openness, agreeableness, and conscientiousness respectively (Costa & McCrae, 1992). There are 60 questions, each answered on a five point likert scale ranging from “strongly agree” to “strongly disagree”.  The scale is a shortened version of the NEO Personality Inventory.
    Participants were then administered the Rosenberg Self-Esteem Scale. This scale is a 10 item self report scale, also with four point likert scoring ranging from SA (Strongly Agree) to SD (Strongly Disagree). The test has high reliability, with test-retest correlations in the range of .82 to .88 (Blascovich and Tomaka, 1993). This study was interested in the overall score of self-esteem, which was found by reversing the questions that were negatively worded and summing these responses with the positively worded questions. (Rosenberg, 1986). The scores that are higher are representative of higher self esteem. For this study, a mean score was found for the results, and those responses above the mean were considered “high” self-esteem, while those below were considered “low”.
        Materials
     Participants used the Sensomotoric Head Mounted Eye Tracking System, which tracked their eye focus during a conversation. The tracking system consists mainly of a helmet that is worn by the participant. This helmet has two cameras, one that shows what the participant is looking at and one that focuses on the eye itself. The system tracks the movement of the pupil, and turns this data into a representative red dot placed on the recording of the camera. In other words, when the experimenter reviews the recording of what the subject was looking at, a red dot will show the direction the eyes are gazing in. This allows the direction of gaze to be tracked and recorded.
     In addition to the headwear, an operator PC was used to run the system that supports the Head Mounted Eye Tracking System. This system, IViewX, records the data from the cameras and produces MPEG and data files to save information. This system was also used to calibrate participants to the Eye Tracker.  The software for the Eye Tracker is owned by SensoMotoric Instruments GmbH and is protected by the Federal Republic of Germany copyright laws. Software for this instrument was made in Teltow, Germany from December 1997-2002.
        Procedures
     The participants took the self-esteem and personality assessments first. Following the test they were fitted with the Eye Tracker, and the machine was calibrated to the participant. A small amount of deception was used by telling the participant the researcher had to leave the room. During the period when the experimenter was gone, a confederate had a general, non-scripted conversation with the participant, while the eye tracker traced the movements of the eyes of the participant. Following this conversation subjects were debriefed, and allowed to leave. The males and females were divided into two groups, one that spoke to a male assistant and one that spoke to a female. This controlled for the gender of the research assistant and any other effects that may have come from using only one research assistant. Each conversation lasted 5 minutes.

Back to top

Results

     The intent of this study was to assess the relationship between self-esteem, personality and gaze in a conversation. Self-esteem was measured through a self-report survey, the Rosenberg Self-Esteem Scale. Scores for this test were tallied by hand, and an overall sum score was used. Personality was measured through the NEO Five Factor Inventory, a personality scale developed by Costa and McCrae. This inventory was also tallied by hand. These scores were compared to the dependent variable of eye gaze in a conversation. Gaze was measured by the number of times a participant looked away from the research assistant throughout a five-minute conversation period. To establish the reliability of the rater, a second rater was used to view the data. A strong positive correlation (r = .86) was established.
    A Pearson Product-moment correlation was used to establish a relationship between the dependent variables. A significant negative relationship was found between Self-esteem scores and eye gaze (r = -.588, p = .017), supporting the hypothesis of this study. As scores for the Rosenberg SES increased, the number of times the participant looked away from the confederate decreased. Participants who looked away less maintained more eye contact with the confederate. However, no significant relationship was found between gaze and any of the five personality factors, failing to fully support the hypothesis. However several of the correlations were approaching significant, and could be considered trends. These coefficients are shown in the below table, and those that are significant are marked.

Table of correlations for self-esteem, eye contact and personality

 Back to top

Discussion
     This study hypothesized a relationship between personality factors, self-esteem and eye contact. That is, it was predicted that the higher a person's self-esteem level, and the higher a person's score for various personality factors, the higher the level of eye contact a person will have in a conversation. Further, it was proposed that males and females would have different levels of eye contact depending on their test scores. The results of the analyses showed an overall relationship between self-esteem and eye contact, which supported the proposed hypothesis. However, no relationship was found between the personality factors tested in the NEO. There was also no evidence of a gender difference in the data.
     The results from this study parallel some of the findings by Lagomarsino, et al. in their 1998 study. These researchers found that participants with high self-esteem, when asked to view two separate videos of two research assistants engaging in a conversation (one with high eye contact and one with low contact) rated the research assistants as having corresponding levels of self-esteem. In other words, these participants rated the research assistant who maintained eye contact throughout the conversation as having high self-esteem, and the assistant who did not maintain eye contact as having low self-esteem. Interestingly, the ratings of the participants did not correspond with their own level of self-esteem; that is, those with high self-esteem did not necessarily rate the person in the study as high in self-esteem even when they kept consistent eye contact through the conversation. In the current study, the results of Lagomarsino's participants are shown to be fairly reliable, but through a different method. Rather than asking the participants to observe another conversation and rate the people in that conversation, these participants were actually involved in a conversation that was observed in order to correlate the gaze of the participant with their scores of self-esteem and personality. This study looked to make a stronger connection between eye contact and self-esteem than the estimation of another. According to the results of the current study, as eye contact increases in conversation, so does self-esteem. Studies have shown the relative inability to correlate self-esteem solidly with scores of personality. Watson, Suls & Haig (2002)noted the lack of previous research to back claims of relationships. However, despite the difficulty in connecting esteem to the general characteristics assessed in the NEO, Watson’s study had some success pairing scores on the Rosenberg Self-Esteem Scale with lower level traits of the NEO. The Five Factor Inventory used for the purposes of this study were not specific enough to make a connection, and future research may want to use a more specific measure for the various traits of personality. Such a decision could allow a stronger or clearer relationship between self-esteem scores and personality.
      The applications of this research could be widespread. However, the methodological setup of this study could have been adjusted in some ways to improve the validity of this research. The ecological validity, that is the amount the setting and procedures of the experiment parallel a realistic situation, is questionable. The participant was required to wear the eye tracker, which could be considered strenuous, because such a situation would not likely occur on a daily basis. Further, the participant was aware that the eye tracker was recording everything that took place from the moment the researcher left the room. Also, because the participant knew the nature of the eye tracker, there is a possibility that some could have guessed the true nature of the experiment. In other words, because the participant was informed of the function of the eye tracker, he or she could have inferred the purpose of the conversation and contaminated the results. The effectiveness of the deception was not examined in this study, and a small deception questionnaire could be used in the future to control for this feature.  Furthermore, some of the participants were acquainted with the research assistants, and because of this were at an advantage. These individuals had a conversation with a person they knew, while others had to speak to a person they had never met before, or maybe knew vaguely. Factors like this could have been controlled for if a larger population of subjects was available for research.
 Also, the research assistants could have been an issue. Of each group of eight males and females, half spoke to a male assistant and half spoke to a female assistant to control for the gender of the assistant. However, due to scheduling and other factors, the same research assistants could not be maintained through the research. The gender was constant, but the same person was not always used. This could be another threat to the validity of the experiment, that could be controlled for in future research. Even the low amount of participants in this study is a small aspect that could make a big difference, especially to the generalizability of the results. Future research could control for the errors in this study, and present a stronger case for the relationship between self-esteem and eye contact.
     In a recreational sense, this information could be extremely helpful to coaches of all ages. Especially at higher levels, where recruiting is involved, athletes have very brief meetings with coaches and players for prospective teams, and this is where an evaluation of self-esteem or personality traits could be extremely useful. If these traits can be implied by levels of eye contact, coaches could have a better grasp on the type of athletes that come to their offices for meetings. Often it is difficult to make a judgment on a person from such a small example of their character, but with this information that assessment could be more accurate, if even just a little.

Back to top

LINKS

 Sensomotoric Instruments
 National Association for Self-Esteem
 Eye Contact

References

Amalfitano, J; Kalt, N. (1977). Effects of eye contact on the evaluation of job applicants. Journal of Employment Counseling, 41 (10-A), 4508-4509.
Argyle, M. (1976). The Psychology of Interpersonal Behaviour. London: Penguin Books.
Argyle, M. & Cook, M. (1976). Gaze & Mutual Gaze. London: Cambridge University Press.
Baron - Cohen, S. (1994). How to build a baby that reads minds: Cognitive mechanisms in mind reading. Cahiers de Psychologie Cognitive, 13.
Baumeister, R; Campbell, J; Krueger, J; Vohs, K. (2003). Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the Public Interest, 4, 1-44.
Blascovich,J.& Tomaka, J. (1993). Measures of Self-Esteem. 115-160.
Brooks, C.; Church, M.; Fraser, L. (1986). Effects of duration of eye contact judgments of personality characteristics. Journal of Social Psychology, 126, 71-78.
Bushman, B; Baumeister, R. (1998). Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of Personality and Social Psychology, 75 (1), 219-229.
Costa, P.T.& McCrae, R.R. (1992). NEO PI-R. Odessa, Florida: Psychological Assessment Resources, Inc.
Crocker, L; Park, L. (2004). The costly pursuit of self-esteem. Psychological Bulletin, 130, 392-414.
Ekman, P. & Friesen, W. (1969). Nonverbal leakage and clues to deception. Psychiatry, 32.
Friesen, C; Kingstone, A. (1998). The eyes have it! Reflexive orienting is triggered by nonpredictive gaze. Psychonomonic Bulletin & Review, 5, 490-495.
Garcia, S; Stinson, L; Ickes, W; Bissonnette, V; Briggs, S. (1991). Shyness and Physical Attractiveness in Mixed Sex Dyads. Journal of Personality and Social Psychology, 61.
Guegen, Nicholas; Jacob, Celine. (2002). Direct look versus evasive glance and compliance with a request. Journal of Social Psychology, 142.
Heron, J. (1970). The Phenomenology of Social Encounter: The Gaze. Philosophy and Phenomenological Research, 31.
Kellerman, J; Lewis, J; Laird, J.D. (1989). Looking and loving: The effects of mutual gaze on feelings of romantic love. Journal of Research in Personality, 23(2), 145-161.
Kendon, A. (1967). Some Functions of Gaze Direction in Social Interaction. Acta Psychologica, 32.
Kleinke, C.L. (1986). Gaze and eye contact: A research review. Psychological Bulletin, 100 (1), 78-100.
Lafrance, M; Mayo, C. (1976). Racial differences in gaze behavior during conversation: Two systematic observational studies. Journal of Personality and Social Psychology, 33(5), 547-552.
Lagomarsino, J; Gallagher, M; Yankalunas, S; Brooks, C; O’Brien, J. (1998). Perception of a model’s self-esteem as a function of observer self-esteem and model’s duration of eye contact. Psychological Record, 48.
Macrae, N; Hood, B; Mason, M; Milne, A; Rowe, A. (2002). Are you looking at me? Eye Gaze and Person Perception. Psychological Science, 13.
Marcelle, M. (1976). Eye contact as a function of race, sex, and distance. Unpublished Doctoral Dissertation.
Morton, J; Johnson, M. (1991). CONSPEC and CONLEARN: A two-process theory of infant face recognition. Psychological Review, 98(2), 164-181.
Perrett, D; Emery, N.J. (1994). Understanding the intentions of others from visual signals: Neuropsychological evidence. Cahiers de Psychologie Cognitive, 13.
Pronin, K; Gilovich, T; Ross, L. (2004). Objectivity in the eye of the beholder: Divergent perceptions of bias in self versus others. Psychological Review, 111(3).
Rosenberg, M. (1986). Conceiving the Self. Florida: Krieger.
Watson, D; Suls, J; Haig, J. (2002). Global self-esteem in relation to structural models of personality and affectivity. Journal of Personality and Social Psychology, 83(1).

Back to top