Technical ReportPDF Available

Gaze Behaviour in Response to Direct and Averted Dynamic Facial Gaze and the Influence of Social Anxiety and Stuttering Position Paper



The position paper analyses research in gaze behaviour in response to direct and averted dynamic facial gaze and the influence of social anxiety and stuttering. The paper shows summaries of research which are both for and against the statement including studies showing eye gaze is one of the most important elements in social interactions, indicating the significance of eye gaze. Each aspect of the topic has been broken down to show its importance, and why gaze behaviours might be influenced by social anxiety and stuttering. The author states gaze behaviour does influence social anxiety and stuttering. The author makes arguments for and against gaze behaviour having an influence on social anxiety and stuttering, but ultimately concludes that it does have an influence.
Gaze Behaviour in Response to Direct and Averted
Dynamic Facial Gaze and the Influence of Social
Anxiety and Stuttering
Position Paper
Gareth V. Walkom
Nottingham Trent University, Nottingham, UK
Abstract The position paper analyses research in gaze
behaviour in response to direct and averted dynamic facial gaze
and the influence of social anxiety and stuttering. The paper
shows summaries of research which are both for and against the
statement including studies showing eye gaze is one of the most
important elements in social interactions, indicating the
significance of eye gaze. Each aspect of the topic has been broken
down to show its importance, and why gaze behaviours might be
influenced by social anxiety and stuttering. The author states
gaze behaviour does influence social anxiety and stuttering. The
author makes arguments for and against gaze behaviour having
an influence on social anxiety and stuttering, but ultimately
concludes that it does have an influence.
Keywords - Gaze direction, eye contact, direct gaze, averted
gaze, social anxiety, stutter, stammer, avoidance, behaviour
The purpose of this position paper is to highlight and
support research indicating gaze behaviour in response to direct
and averted dynamic facial gaze and the influence of social
anxiety and stuttering. It has been shown in social interactions
that gaze cues provide information, regulate interaction,
express intimacy, exercise social control, and facilitate goal
setting (Gamer, 2007). Hietanen et al. (2008) states eye gaze is
a powerful stimulus in social interaction. Seeing another person
looking at you is likely to indicate that he or she is attending to
you, whereas another person’s gaze averted away from you
indicates that his or her attention is directed to somewhere else.
(Hietanen et al. 2008). Gaze behaviour is a crucial factor in
defining relationships and in allowing reciprocal influences to
be exchanged as persons relate (Grumet, 1983). Eye contact is
a socially facilitative behaviour that influences how attractive,
attentive, socially competent, credible, and socially dominant
one will be perceived by others (Weeks et al. 2013). Overall
findings from existing studies indicate that normative eye
contact helps navigate social interactions (Langer et al. 2016).
Eye contact is also an important part of stuttering therapy,
as Atkins, (1988) states the development of effective eye
contact needs to be incorporated as a major goal in stuttering
therapy, following the fact stutterers tend to avoid eye contact
when talking to others. Direct gaze may be a fear-relevant
feature for socially anxious individuals in social interactions
(Wieser et al. 2008). However, Wieser et al. (2008) states this
seems not to result in gaze avoidance. This paper examines
current literature to conclude with a position to whether social
anxiety and stuttering influences gaze behaviour in response to
direct and averted facial gaze.
A. Direct and Averted Facial Gaze
People are faster to detect a face with a direct than averted
gaze, and direct gaze facilitates facial gender discrimination
and identity recognition (Helminen et al. 2011). Even new-born
infants (2 days old) can discriminate between straight and
averted gaze suggesting that the ability to detect gaze direction
is innate (Hietanen et al. 2008). Hietanen et al. (2008) also
indicates that direct gaze is more efficiently relative to averted
B. Social Anxiety
Social anxiety disorder (SAD) is one of the most common
psychiatric disorders, with a lifetime prevalence of 5% to 13%,
and it frequently causes severe distress and impairment
(Schneier et al. 2011). Schneier et al. (2011) also say that
common measures of severity of social anxiety quantify fear
and avoidance across a variety of social or public situations or
focus on severity of anxious states and interference with
functioning in social interactions (e.g. Social Interaction
Anxiety Scale [SIAS]). The SIAS is a set of 20 characteristics,
where the participant must rate each characteristic as either: not
at all, slightly, moderately, very, or extremely. This measures
SAD by assessing the distress of the participant when meeting
and talking with others. This scale and many other SAD scales
are also used in numerous clinical settings and research.
C. Stuttering
Stuttering is a speech motor disorder that involves
disruptions to the free flow of speech production. The
condition is understood to be the result of neural processing
deficits, impairing the initiation of speech motor programmes
for the production of syllables (Lowe et al. (2012).
In the diagnosis of individuals who stutter, there are many
different aspects of speech fluency that can be examined, some
examples of stuttering behaviours that may be considered
include: the frequency, duration, types, and severity of
disfluencies (Yaruss, 1998). It has been proven that stuttering
affects four times as many males as it does females and people
do not generally stutter when they sing, whisper, speak in
chorus, or when they do not hear their voice (City Lit, 2010).
The importance of gaze has been highlighted in many
papers to date. Many of them indicate how important eye gaze
is while engaging in a conversation. Rutter, (1984) expresses
how mutual gaze is an essential component of dyadic face-to-
face interaction and is often used to establish social
interactions. The enhanced affective arousal elicited by the eye
contact is supposed to influence subsequent perceptual and
cognitive processing (Senju, 2009). Eye gaze is one of the most
important elements in human-human interactions: we use our
eyes to study the behaviour and appearance of others (Hugot,
2007). Without direct gaze during social interactions, the
listener may not know that the person who is speaking is
talking to them, showing eye contact is important when
interacting with others in conversation. It is also important to
understand if social anxiety and stuttering does influence gaze
behaviour, as this could indicate gaze needs to be analysed in
future studies in social anxiety and stuttering.
The author suggests gaze behaviour in response to direct
and averted dynamic facial gaze is influenced by social anxiety
and stuttering. The amount of research on the opposing side is
minimal and lacks research justifying the findings. A study by
Mühlberger et al. (2008) has results which indicate that faces
are initially avoided in socially anxious participants. This
shows that not only do individuals with SAD avoid eye contact
within conversations, but the facial expressions of the person
they are talking to (happy or angry) may also have an impact
on levels of anxiety in SAD. These results suggest that happy
or angry faces could make an individual feel different
depending on the facial expression. This could also lead to gaze
avoidance in those with social anxiety if their anxiety levels are
increased when presented with an angry face. Howell et al.
(2015) create findings from their study which provide further
support that individuals with elevated social anxiety are likely
to engage in eye contact avoidance, which may have negative
impacts on social functioning and contribute to the
maintenance of social anxiety symptoms.
Some socially anxious individuals will develop attentional
strategies, which generally include avoiding the face of an
individual, especially mutual gaze. Horley et al. (2001) show
very relevant results to this, where the study suggests that
attentional strategies for the active avoidance of salient features
are an important marker of interpersonal cues of social phobia.
The salient features observed in the study were the eyes, nose,
and mouth, which produced scan paths that represented and
inverted triangle shape to achieve their results. Chen et al.
(2001) has findings which are also in line with this, showing
patients with generalised social phobia directed their attention
away from all faces, regardless of emotion expression.
Individuals with social anxiety may also avoid eye contact as a
fear of being evaluated. This is a common avoidance
behaviour, which can sometimes be shown when individuals
with social anxiety have social encounters. Schulze et al.
(2013) summarise with avoidance of eye contact in social
anxiety may be understood as an attempt to avoid signs of
social threat and to regulate excessive fears of being evaluated.
This may be an indication of a safeguard behaviour, where an
individual with social anxiety will avoid certain social
situations they know will be difficult to manage. The results
could also suggest excessive fears of being evaluated are
common in those with social anxiety, meaning gaze avoidance
would be more common in those situations.
Most methods of stuttering therapy evaluate eye gaze and gaze
behaviours during speech, especially during moments of a
stutter, as some People Who Stutter (PWS) can close their
eyes, flicker their eyes, or lose eye contact during moments of a
stutter. A study by Hudock et al. (2015) indicates gaze aversion
from the other person in the conversation can also affect PWS
when speaking and may contribute to their negative reactions
to their own stuttering. This study suggests a build-up of
negative reactions and thoughts could increase over time,
which could increase an individual’s fear of disfluency. As
gaze avoidance is common with PWS, negative reactions and
thoughts could increase the avoidance of gaze.
On an anecdotal note, the author also has a stutter, and they
have experienced eye flickers, the closing of the eyes and gaze
avoidance while in moments of stuttering in social interactions.
This has led to the author noticing the listener losing eye
contact with the author, and in some accounts losing interest in
the conversation. The author has also received stuttering
therapy, which introduced gaze behaviour while stuttering. Of
which, the author noticed better control of their stutter while
engaging in a conversation with better gaze behaviour.
It has been shown that self-reported fear and avoidance of
eye contact are associated with social anxiety in both
nonpatient and SAD samples (Schneier et al. 2011). This
research implies those without SAD also have a certain level of
gaze avoidance, suggesting gaze behaviour may not only be
influenced by social anxiety.
A study by Wieser et al. (2008) considers the effects of
direct and averted gaze on autonomic arousal and gaze
behaviour in social anxiety. The study results indicate the
groups they tested did not differ in their gaze behaviour
concerning direct vs. averted gaze. Their psychological
findings indicate that direct gaze may be a fear-relevant feature
for socially anxious individuals in social interaction. However,
this seems not to result in gaze avoidance.
They study by Schneier et al. (2011) had a few limitations.
As Schneier, et al, (2011) developed the Gaze Anxiety Rating
Scale (GARS), it should be taken into consideration that this
scale had not been validated by others before, meaning it may
not be 100% reliable. The research also lacked in samples to
conduct factor analysis, and the research also needed more
stringent testing of divergent validity with depression.
Schneier, et al, (2011) may have had nonpatients showing signs
of self-reported fear and eye contact due to bad analysis of
participants during the assessment stage. This could also be due
to participants giving false results.
From the study by Wieser et al. (2008), it was said research
on gaze processing in social anxiety should be intensified and
the impact of direct gaze on perceived anxiety should be
further investigated as this might be a critical threat stimulus
and result in avoidance behaviour like gaze avoidance. This
study goes against most other studies on gaze behaviour, as it
implies gaze behaviour concerning direct vs. averted gaze does
not differ, and gaze avoidance is not a fear-relevant feature in
Overall, it has been shown throughout this position paper
that direct gaze does result in gaze avoidance in socially
anxious individuals and stuttering.
There is only a small minority of research that suggests this
statement is false, which does have many limitations to it.
There has however been a lot of research on the topic since
then, which shows direct gaze does result in gaze avoidance in
socially anxious individuals and stuttering.
Nearly all research papers to date summarise gaze
behaviour does play a part within SAD. This has been proven
by indicating avoidance in eye gaze during a conversation,
which is more visible in SAD. As stuttering symptoms are
similar to SAD symptoms, there is evidence that results should
be similar. Studies in stuttering and gaze behaviours back this,
showing gaze behaviour is different with PWS compared to
people who do not stutter.
This position paper has demonstrated that gaze behaviour
needs to be considered and analysed in future studies in social
anxiety and stuttering, as it could have a key impact on the
Atkins, C. P. (1988). Perceptions of speakers with minimal eye contact:
Implications for stutterers. Journal of Fluency Disorders, 13(6), 429-
City Lit. (2010). Stammering Therapy An Intergrated Approach, 1, London:
City Lit.
Gamer, M., Hecht, H. (2007). Are You Looking at Me? Measuring the Cone
of Gaze. Journal of Experimental Psychology: Human Perception and
Performance, 33(3) (2007), 705-715.
Grumet, G. W. (1983). Eye contact: the core of interpersonal relatedness.
Psychiatry, 46(2), 172-180.
Helminen, T. M., Kaasinen, S. M., Hietanen, J. K. (2011). Eye contact and
arousal: The effects of stimulus duration. Biological Psychology, 88
(2011), 124-130.
Hietanen, J. K., Leppänen, J. N., Peltola, M. J., Linna-aho, K., Ruuhiala, H. J.
(2008). Seeing direct and averted gaze activates the approach-avoidance
motivational brain systems: Neuropsychologia, 46 (2008), 2423-2430.
Horley, K., Williams, L. M., Gonsalvez, C., Gordon, E. (2001). Social phobics
do not see eye to eye: A visual scanpath study of emotional expression
processing. Anxiety Disorders, 17 (2003), 33-44.
Howell, A. N., Zibulsky, D. A., Srivastav, A., Weeks, J. W. (2015). Relations
among Social Anxiety, Eye Contact Avoidance, State Anxiety, and
Perception of Interaction Performance during a Live Conversation.
Cognitive Bahaviour Therapy, 45(2), 111-122.
Hudock, D., Stuart, A., Saltuklaroglu, T., Zhang, J., Murray, N., Kalinowski,
J., Altieri, N. (2015). Segmented Analysis of Eye Gaze Behaviors of
Fluent and Stuttered Speech. Canadian Journal of Speech-Language
Pathology and Audiology, 39(2) (2015), 134-145.
Hugot, V. (2007). Eye Gaze Analysis in Human-Human Interactions.
Master’s Thesis in Speech Communication. Royal Institute of
Technology, Stockholm, Sweden.
Langer, J. K., Lim, M. H., Fernandez, K. C., Thomas, L. R. (2016). Social
Anxiety Disorder is Associated with Reduced Eye Contact During
Conversation Primed for Conflict. Cognitive Therapy and Research.
Lowe, R., Guastella. A. J., Chen, N. T. M., Menzies, R. G., Packman, A.,
O’Brian, S., Onslow, M. (2012). Avoidance of eye gaze by adults who
stutter. Journal of Fluency Disorders, 37 (2012), 263-274.
Mühlberger, A., Wieser, A., Pauli, P. (2008). Visual Attention during Virtual
Social Situations Depends on Social Anxiety. CyberPsychology &
Behavior, 11(4), 425-430.
Rutter, D. R. (1984). Looking and seeing: The role of visual communication in
social interaction. Chichester, UK: Wiley.
Schneier, F. R., Rodebaugh, T. L., Blanco, C., Lewin, H., Liebowitz, M. R.
(2011). Fear and avoidance of eye contact in social anxiety disorder.
Comprehensive Psychiatry, 52 (2011), 81-87.
Schulze, L., Renneberg, B., Lobmaier, J. S. (2013). Gaze perception in social
anxiety and social anxiety disorder. Frontiers in Human Neuroscience
7(872) (2013), 1-4.
Senju, A., Johnson, M. H. (2009). The eye contact effect: mechanisms and
development. Trends in Cognitive Sciences, 13, 127-134.
Weeks, J. W., Howell, A. N., Goldin, P. R. (2013). Gaze Avoidance In Social
Anxiety Disorder. Depression and Anxiety, 00:1-8 (2013).
Wieser, M. J., Pauli, P., Alpers, G. W., Mühlberger, A. (2008). Is eye to eye
contact really threatening and avoided in social anxiety? An eye-
tracking and psychophysiology study. Journal of Anxiety Disorders, 23
(2009), 93-103.
Yaruss, J. S. (1998). Real-Time Analysis of Speech Fluency: Procedures and
Realiability Training. American Journal of Speech-Language Pathology,
7(2), 25-37.
... Throughout the process, a Galvanic Skin Response sensor was attached to the fingers of a user so that we can monitor how much stress he is under [1]. EDA anxiety levels have been found to increase over course of time spent in VR therapy due to exposure [20]. Machine Learning will utilize the averaged stress level over a five second window in the preparatory relaxation scene to determine the level of difficulty level in the speech training scene, just as we do with voice input. ...
... Another technology that would help with determination of user behavior is eye tracking during anxious moments. Eye flicker, loss-of-contact, and avoidance of gaze is particularly prominent during and before moments of stutter [20]. By detecting gaze during training and testing sessions in VR, we can gauge not only when speakers are anxious, but we can also infer what they are anxious about. ...
Full-text available
Eye contact is important for successful social interactions (e.g., Dalton et al. in Nat Neurosci 8:519–526, 2005. doi:10.1038/nn1421), suggesting that gaze avoidance could be damaging for social functioning. Gaze avoidance has been proposed to relate to higher social anxiety (Schneier et al. in Compr Psychiatry 52:81–87, 2011. doi:10.1016/j.comppsych.2010.04.006), yet studies utilizing behavioral observation have produced mixed findings (Farabee et al. in J Res Personal 27:365–376, 1993. doi: 10.1006/jrpe.1993.1025; Walters and Hope in Behav Ther. 29:387–407, 1998. doi:10.1016/S0005-7894(98)80039-7; Weeks et al. in J Soc Clin Psychol 30:217–249, 2011. doi:10.1521/jscp.2011.30.3.217). The goal of the current study was to clarify the mixed findings in the literature utilizing a clinical sample. Participants completed interactions with another participant. We assessed eye contact using independent coding. Participants with social anxiety disorder made lower levels of eye contact relative to participants without social anxiety disorder during a conversation primed for conflict. Integrating these findings with previous work, we theorize that social anxiety relates to eye contact when there is an impairing level of social anxiety and the interaction is primed for conflict.
Full-text available
There is building evidence that highly socially anxious (HSA) individuals frequently avoid making eye contact, which may contribute to less meaningful social interactions and maintenance of social anxiety symptoms. However, research to date is lacking in ecological validity due to the usage of either static or pre-recorded facial stimuli or subjective coding of eye contact. The current study examined the relationships among trait social anxiety, eye contact avoidance, state anxiety, and participants' self-perceptions of interaction performance during a live, four-minute conversation with a confederate via webcam, and while being covertly eye-tracked. Participants included undergraduate women who conversed with same-sex confederates. Results indicated that trait social anxiety was inversely related to eye contact duration and frequency averaged across the four minutes, and positively related to state social anxiety and negative self-ratings. In addition, greater anticipatory state anxiety was associated with reduced eye contact throughout the first minute of the conversation. Eye contact was not related to post-task state anxiety or self-perception of poor performance; although, trends emerged in which these relations may be positive for HSA individuals. The current findings provide enhanced support for the notion that eye contact avoidance is an important feature of social anxiety.
Full-text available
Purpose: To measure the effect of stuttering on eye-gaze in fluent speakers while viewing video presentations of typical speakers and people who stutter (PWS) speaking because eye-gaze behaviors provide indicators of emotion and communicative integrity. Method: Sixteen fluent college-age adults, naïve to stuttering, observed six 30-second audiovisual speech samples of three PWS, and three age and gender matched controls who do not stutter (PWNS). A desk-mounted eye-tracker recorded the amount of time participants spent watching four regions of interest (ROIs) in the stimulus videos of PWS and PWNS: eyes, nose, mouth, and " outside " (i.e., any gaze-point not occurring within the eyes, nose, or mouth area). Proportions of gaze-time in each ROI were the dependent variables of interest in the study. Comparisons were made between proportions of time spent in each ROI for the PWS and PWNS speaker groups, and also between fluent versus disfluent speech segments produced by the PWS. Results: Participants spent significantly more time watching the eyes (e.g., maintaining eye-contact) when viewing PWNS than PWS. They also spent significantly more time observing mouth regions of PWS. When watching the videos of PWS, participants spent significantly more time observing nose and mouth regions when speech was stuttered (PWS-S) than when the speech was fluent (PWS-F). Conclusions: Overall, the difference in eye gaze patterns across speaker-group is interpreted to indicate negative emotional responses to stuttering. Current findings align with previous research showing that stuttered speech elicits negative reactions from listeners. Specifically, stuttering behaviors avert gaze from the eyes. Gaze aversion is a clear sign of disrupted communication that is visible to PWS and may contribute to their negative reactions to their own stuttering. Analyse segmentée des comportements de fixation du regard sur une élocution fluide et une élocution bégayée
Full-text available
Clinical observations suggest abnormal gaze perception to be an important indicator of social anxiety disorder (SAD). Experimental research has yet paid relatively little attention to the study of gaze perception in SAD. In this article we first discuss gaze perception in healthy human beings before reviewing self-referential and threat-related biases of gaze perception in clinical and non-clinical socially anxious samples. Relative to controls, socially anxious individuals exhibit an enhanced self-directed perception of gaze directions and demonstrate a pronounced fear of direct eye contact, though findings are less consistent regarding the avoidance of mutual gaze in SAD. Prospects for future research and clinical implications are discussed.
Full-text available
The relationship between gaze avoidance and social anxiety has been examined previously using eye-tracking and static social images. Overall, findings to date highlight increased gaze avoidance as a behavioral marker of social anxiety. The purpose of the present study was to better elucidate the relationship between gaze avoidance and social anxiety disorder (SAD) symptoms via covert eye tracking of gaze tendencies in response to a dynamic computerized social interaction simulation. On the basis of the bivalent fear of evaluation (BFOE) model of social anxiety,([1]) it was expected that participants with SAD, compared to nonsocially anxious control (NSAC) participants, would exhibit gaze avoidance in response to both positive and negative social feedback. Participants with SAD (n = 20), and a sample of demographically equivalent NSAC (n = 19), were administered clinical diagnostic interviews and a computerized social simulation task. The simulation task consisted of viewing 26 dynamic videos (13 positive and 13 negative), each 12 s in duration. All participants were covertly eye tracked during the simulation. SAD participants exhibited greater global gaze avoidance in response to both the positive and negative video clips in comparison to the controls. Moreover, the SAD group exhibited equivalent gaze avoidance in response to stimuli of both emotional valences. These results provide additional support for gaze avoidance as a behavioral marker of SAD, as well as additional support for the BFOE model. Implications for the assessment of SAD are discussed.
Full-text available
Purpose: Adults who stutter are at significant risk of developing social phobia. Cognitive theorists argue that a critical factor maintaining social anxiety is avoidance of social information. This avoidance may impair access to positive feedback from social encounters that could disconfirm fears and negative beliefs. Adults who stutter are known to engage in avoidance behaviours, and may neglect positive social information. This study investigated the gaze behaviour of adults who stutter whilst giving a speech. Method: 16 adults who stutter and 16 matched controls delivered a 3-min speech to a television display of a pre-recorded lecture theatre audience. Participants were told the audience was watching them live from another room. Audience members were trained to display positive, negative and neutral expressions. Participant eye movement was recorded with an eye-tracker. Results: There was a significant difference between the stuttering and control participants for fixation duration and fixation count towards an audience display. In particular, the stuttering participants, compared to controls, looked for shorter time at positive audience members than at negative and neutral audience members and the background. Conclusions: Adults who stutter may neglect positive social cues within social situations that could serve to disconfirm negative beliefs and fears. Educational objectives: The reader will be able to: (a) describe the nature of anxiety experienced by adults who stutter; (b) identify the most common anxiety condition among adults who stutter; (c) understand how information processing biases and the use of safety behaviours contribute to the maintenance of social anxiety; (d) describe how avoiding social information may contribute to the maintenance of social anxiety in people who stutter; and (e) describe the clinical implications of avoidance of social information in people who stutter.
Many authors have suggested that it is possible for clinicians to collect basic data regarding their client's speech fluency on-line, or in real time, while the client is speaking. Unfortunately, the literature contains relatively little in the way of detailed instructions on exactly how such data should be collected. This article provides specific instructions for real-time collection of information about the frequency and types of speech disfluencies produced by individuals who stutter. The paper also outlines procedures for training students and clinicians to use this technique reliably and accurately and proposes tolerance limits for determining whether frequency counts are sufficiently reliable for clinical use.
To investigate the role of visual communication in social interaction 2 experiments were conducted with blind (Exp I) and with sighted (Exp II) Ss. In Exp I, physical presence was varied. In Exp II, Ss conversed from separate rooms over a video link so that they were physically separated but visually together (Condition 1), or a curtain was partially pulled across the room between the Ss, so that they were physically together in the same room but visually separated (Condition 2). Results indicate that (a) visual communication is less important in social interaction than has generally been supposed, and (b) what is important, for a variety of measures of content, style, and outcome, is the total number of social cues the participants have available from visual communication, physical presence, and other sources. (9 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
One hundred and thirty-three college students responded to percentage definitions of “good,” “minimal,” and “no” eye contact. They also judged their perceptions of a speaker with “little or no eye contact” on a semantic differential scale containing 60 polarized adjective pairs regarding personality traits. The majority of the student respondents defined an individual with “good” eye contact as looking at his/her audience/listener 90–100% of the time. This is a more stringent definition than other literature has indicated. Furthermore, speakers with little or no eye contact were judged less favorably on 70% of the items, which indicates that low eye contact adversely affects perceptions of a speaker's personality. Therefore, stutterers need to be made aware of the importance of and determinants of effective eye contact in communication. Also, the development of effective eye contact needs to be incorporated as a major goal in stuttering therapy.
The present study investigated the effect of stimulus duration on skin conductance responses (SCRs) evoked by different gaze directions of a live person. In two separate parts of the experiment, either two fixed stimulus durations (2s and 5s) or a participant-controlled stimulus duration was used. The results showed that the eye contact evoked enhanced SCRs compared to averted gaze or closed eyes conditions irrespective of the presentation time. Subjective evaluations of approach-avoidance-tendencies indicated that the direct gaze elicited either approach or avoidance, depending on the participant. Participants who had evaluated a direct gaze-condition as approachable were found to be more emotionally stabile than those who had evaluated the same condition as avoidable. In the self-timing condition, averted gaze was looked at longer than direct gaze. Our results suggest that direct gaze, also when encountered only briefly like in every-day social encounterings, increases autonomic sympathetic arousal.