Correlations between variables involved in the proposed model
1.1. Social anxiety
Social anxiety refers to the excessive and persistent fear that a person will be embarrassed and/or rejected by other people in one or more social or performance situations . When a socially anxious person wants to present a desirable image of him/herself, a strong desire to accomplish it is accompanied by considerable uncertainty if he or she can really do it. Almost every person has to some extent felt socially anxious (feeling weird, blushing or stammer) in some social situations or situations in which he or she has been evaluated.
People differ in how often and with what intensity they feel socially anxious and show a certain degree of consistency in how anxious they are across social situations and over time . Thus, some people are by nature strongly and more frequently socially anxious then other and we are considering this as a feature of the personality. Leary and Kowalski  find that, despite different results, there is little reason for assuming that social phobia/social anxiety disorder is qualitatively distinct from social anxiety as personality trait. Socially phobic persons experience a stronger intensity of anxiety in social situations, their attempts to escape unpleasant social contacts are more extreme and anxiety they are experiencing seriously affects their everyday life. A number of studies show that social anxiety is a continuum, from complete lack of social fear, through the usual forms of shyness and mild social anxiety, to social fears that significantly impair functioning and lead to social anxiety disorder . When social situations become extremely unpleasant to a person and he or she starts to avoid them significantly impairing his/her quality of life, then we are talking of social anxiety disorder .
The fact that social phobia often precedes and/or occurs in comorbidity with other psychiatric disorders  and is often inadequately treated, stresses the need for further research and development of more efficient treatment [5, 6, 7]. When a person seeks professional help for many years after the social anxiety disorder has developed, it is possible that it has been primarily done because of symptoms related to other disorder/s.
Social anxiety disorder is a complex construct with which we try to describe a very heterogeneous group of people. It is important to distinguish people who are afraid of all or almost all social situations, from those who are afraid of only few of them. Likewise, it remains unclear whether the diagnostic subtypes are quantitatively different or reflect different qualitative entities of social phobia.
Studies have confirmed that at least two types of social anxiety should be distinguished: anxiety related to social interaction and anxiety related to some action performance. Anxiety related to social interaction consists of fears when meeting other people (e.g. initiating and maintaining conversation with other people, either in dyads or in groups), while performance anxiety relates to social evaluation concerns of doing something in front of other people (e.g. writing, playing an instrument, giving speeches) which the person would not be afraid doing if alone .
The role of cognitive processes and processes of focusing attention in the maintenance of social phobia are emphasized in contemporary theories [2, 9, 10]. According to these theories, extremely high standards of behavior in social environment maintain social phobia. These standards are characterized by negative thoughts, respectively statements of a person talking to him/herself and his/her assumption that other people see him/her as unsuitable, such as boring [9, 10]. In addition, socially anxious people consider beliefs and assumptions other people have about them as accurate and true. Consequences of these beliefs and assumptions are frequent negative self-statements, negative appraisal of their own behavior in social setting, increased self-focused attention on what has been done wrong in social interaction instead of focusing attention on those aspects which have been done well. Socially anxious people than become preoccupied with thoughts of how they would be evaluated by others and strongly shift attention to detailed monitoring and observation of themselves and impression they leave, including physiological symptoms of anxiety [9, 10, 11].
According to Clark and Wells [9, 11, 12, 13], when social phobics enter a feared situation, a set of assumptions (about themselves and their social world) is activated. If a person estimates situation as dangerous, so-called “anxiety program” is activated. The anxiety program includes physiological, cognitive, affective and behavioral changes that were designed to protect us from harm, but when the danger is largely overestimated this program loses its useful function. Anxiety symptoms, together with the strategies of coping with it, a person can misinterpret as sources of risk, which leads to an exacerbation of anxiety and a series of vicious circles that maintain social anxiety.
Several processes are important in development and maintenance of social anxiety . One of the most important processes is the one concerning the self-focused attention. The shift in attentional focus happens to every social phobic – from perceiving the outside world to detailed monitoring and observation of oneself. The result is the construction of the self as a social object which helps creating an impression of how they appear to others. Onwards, entering feared situation, social phobics tend to use a wide range of safety behaviors they believe are helping them to prevent social disaster. Not only these behaviors are not helpful, they are also very harmful for social phobics. On one hand, safety behaviors can enhance feared behaviors and on the other, it prevents the person from perceiving any positive social feedback which might help in changing unrealistic beliefs. This means that unrealistic beliefs about feared behaviors or the consequences of these behaviors cannot be rejected. According to Clark and Wells, anticipatory and post event processing is very important; it contributes to the instability of self-image of social phobic person and in maintaining a high level of self-focused attention . The basic feature of the anticipatory and post event processes is negativistic thinking concerning the extensive rumination about future failures or real or imagined past failures .
Besides importance of cognitive processes in developing and maintaining social anxiety disorder, it is also important to consider processes of emotion regulation. Many studies show that difficulties with emotion regulation are associated with psychopathology (e.g. 15, 16). It is necessary to explore the role of different ways of emotion regulation in development of specific disorders and to create models which integrate both cognitive and emotion regulation processes in development of psychopathology.
1.2. Emotion regulation
Through emotions we give other people information about our internal condition and behavioral intentions  and therefore they play an important role in interpersonal communication and our lives. They are manifested through specific cognitive, behavioral and physiological responses and are basis for adapting to new situations. If the person assess the situation as relevant to his or her goals and find it as an interesting one, then the emotions start to occur. Someone’s goals may differ in many ways . Goals may be enduring or temporary, conscious and complicated or unconscious and simple. They may be widely shared and understood or highly idiosyncratic. They may have a central role in understanding of ourselves or be peripheral. The base for developing an emotion is the meaning that a person gives to the situation. As the meaning changes over time, the emotion changes, too. Changes in emotional response can be triggered by situational changes or by the changes of significance that the situation has for the person .
Emotions occupy the whole body and include changes in the domains of subjective experience, behavior and physiological reactions. Impulses that encourage us to behave in certain ways and not act otherwise are associated with changes in the autonomic system and neuroendocrine changes. Those changes are followed by a particular behavior . Because of a series of changes in various systems, emotions have imperative quality which means they can terminate the current activity and force a person to become aware of them. Also, when they occur, emotions often have to contend with other reactions that result from the same social context from which they have emerged. This is the most important fact for emotion regulation analysis because of the possibility to modulate emotions in many different ways.
Under certain circumstances, emotions have adaptive function; they have an evolutionary, a social and communicative and a decision making function  but they can also become a source of dysfunction and be maladaptive. It is a challenge to find the way how to regulate our own emotions in order to retain emotional useful features, and to limit their damaging aspect. Emotions can hurt us if they occur at the wrong time or with wrong intensity . The ability to successfully regulate emotions is very important because those inappropriate emotional responses are involved in many forms of psychopathology or in somatic illness development.
A process model of emotion regulation supposes that specific strategies of emotion regulation can vary over a sequence of development of emotional responses [19- 21]. This conception of emotion regulation implies that emotion emerges with an evaluation of emotion cues that can be either external or internal. When a person pays attention to these cues and evaluate them in a certain way, the emotional cues trigger a coordinated set of response tendencies involving experiential, behavioral and physiological systems. Once these reactions occur, they can be modulated in different ways. As emotion develops over a certain period of time, emotion regulation strategies may differ by the point in the emotion-generative process at which they have their primary impact.
A process model of emotion regulation [19, 21- 23] highlights five families of emotion regulation strategies: situation selection, situation modification, attentional deployment, cognitive change and response modulation. Processes related to first four families are considered as
There are clear individual differences in preferred ways of emotional regulation which is important in predicting the behavior of other people . It is shown that people who react better on life demands are the one able to recognize their own emotional states, to understand the meaning of emotions and use their informational value, as well as to adjust the expression of emotion and their own response in a way that fits the context of the situation. This set of abilities is often called emotional intelligence .
Suppression is one of the most widely studied emotion regulation strategies. It is a response-focused emotion regulation and refers to attempts of ignoring already generated emotions and avoiding their expression . Suppression is a way to regulate emotions after cognitive reappraisal of emotional content, respectively it comes relatively late in the emotion-generative process after the behavioral tendencies have been initiated [19- 22]. Studies have shown that this way of emotion regulation is counterproductive because it actually leads to a paradoxical reinforcement of physiological arousal and unwanted affect itself . Suppression of negative emotions brings no relief in sense of its subjective experience . It also leads to decreased expression of both positive and negative emotions and is considered to interfere with relationships triggering unpleasant reactions in other people . This emotional regulation strategy is associated with rare experience of positive emotions and their seldom expression [21, 24, 27].
Although suppression is generally considered to be a maladaptive emotion regulation strategy, it can be adaptive in situations where revealing emotions (e.g. anger or anxiety) should be restrained  or optimum distance between people should be maintained in order to facilitate a smooth social interaction .
1.3. Beliefs about expressing emotions
The way a person will regulate her or his emotion is strongly affected by beliefs she or he has about emotions . If a person does not believe that efforts to regulate emotions will be successful, she or he will consider her/himself incompetent and uncertain and will invest a little effort and energy into implementing strategies of emotional regulation. In contrast, people who believe that emotions can be changed and controlled will be effective in regulating emotions using different adaptive strategies. Beliefs about emotions and emotional expression mediate the relation between experiencing emotions and there expression and thus have impact on emotion-generative process.
Negative reactivity to emotions refers to negative beliefs a person has about emotions, such as fearing consequences following emotion . This construct applies to discomfort when experiencing emotions which leads to strong beliefs that emotional responses are dangerous and harmful for a person.
It has been assumed that socially anxious individuals may refrain from expressing their own emotions to avoid potential rejection. Refraining from expressing emotions offers less “material” for observation, which may cause rejection by others. Studies have shown that socially anxious people indicate a stronger suppression of their emotional experiences, they have lower capacity to monitor, differentiate and describe their own emotions and have more fears related to the experience of emotion and loss of control over them . Spokas, Luterek and Heimberg  have found that beliefs about expressing emotions are significant mediators in the relationship between social anxiety and suppression of emotion, after controlling the effect of social phobics' ability to describe their own emotions to other people and their capacity to monitor them.
Tamir et al.  have confirmed that people who believe emotions are adaptable and changeable shape their own emotions by changing the evaluation of events that caused them. Regardless of the beliefs about emotions people have, they have an equal probability of masking their own feelings in certain situation. Those who believe in the malleability of emotions do not have fixed habit to use suppression as emotional regulation strategy.
1.4. Social anxiety, experiencing positive emotions and quality of life
Although inconsistent, positive emotions can have a lasting impact on our functioning through improvement of our well-being and relations with other people . Research shows that induced positive emotions increase the personal feeling of unity with a close person and increase the confidence that we have in acquaintances. Likewise, the experience of positive emotions expands our attention and reflection in the field of personal and interpersonal functioning.
Other people and interaction with them are the source of positive events and emotions and therefore social activities and a sense of connection with other people are very important for our well-being . There are also clear social benefits from sharing pleasant social events with other people, as they can be attributed to the relationship itself and thus reinforce social ties .
Social phobics are overly focused on the negative outcomes which interfere with their ability to recognize and respond to the potential rewards that come from the environment. It is expected that they experience high levels of negative affect and very low level of positive affect when anticipating participation, participating or constantly thinking about participating in social situation .
The current models of anxiety and depression (e.g., 37) generally assume that only depression is associated with deficits in positive emotions and events. Recent studies show that this deficit is also associated with social anxiety (e.g. 3, 36, 38). Socially anxious people have decreased positive affect and other positive psychological experiences (e.g., curiosity), even after controlling depressive symptoms, and have less frequent and less intense emotional response to positive social events . They report about experiencing less frequent daily positive emotions and events than nonanxious people, and it could not be attributed to the conceptual overlap of social anxiety and other negative affective states . The results have also shown that social phobics reported less positive events experienced during those days when they experienced higher levels of social anxiety and when tended to suppress emotions.
There is a strong evidence of correlation between social anxiety and reduced positive experience . Social anxiety explained an additional 4-5% of variance in positive experiences, after controlling for depression, which is important in understanding this relationship. In his meta-analysis a stable inverse relationship between social anxiety and positive affect has been found (r=-.36; 95% confidence limits (CI): -.31 to -.40) and it remains even after the variance attributed to depressive symptoms and disorders is removed.
It has been found that especially those aspects related to social interaction are related to low positive affect . The significant and negative association of anxiety related to social interaction with all domains of positive psychological functioning, after controlling neuroticism, has been found , while anxiety and fear of being observed by others did not show significant association with these domains.
Social anxiety as a trait is negatively correlated with daily episodes of happiness, relaxation, and positive emotions in general and positively correlated with anger . Results confirmed diminished experience of positive emotions and increased experience of anger in individuals with relatively high levels of social anxiety regardless of being alone or with other people. The authors believe that these two emotional experiences are potentially relevant to socially anxious people.
Socially anxious people express less positive emotions, overall pay less attention to their emotions and have more difficulty in describing their emotions than those with generalized anxiety disorder and control non-anxious group . They express greater fear of anxiety, sadness, anger and even of positive emotions then control group. Insufficient attention to emotions or their frequent ignoring can contribute to difficulties that social phobics have in raising awareness and recognizing their own emotions and in understanding why they feel the way they do. Individuals who are able to recognize and use their emotions are better prepared to flexibly and adaptively respond to environmental requirements and appropriately regulate their affect .
Further studies of relationship between social anxiety and positive emotions are needed. It is well known that positive emotions induce more rapid recovery from adverse physiological effects of negative emotions, increase awareness during activity, efficacy and quality in decision making process and access to more creative and more flexible options in a particular situation . Thus they have impact in life quality which has been found to be impaired in socially anxious people.
In order to understand better the relationship between social anxiety and experiencing positive and negative emotions and life satisfaction in general, the new model has been proposed and tested. Based on the model of social phobia , which emphasizes the role of cognitions, and process model of emotion regulation , especially the response modulation, proposed and tested model has included relationship between social anxiety (two dimensions: general fears and avoidance behaviors concerning social interactions and social evaluation concerns/anxiety related to being observed by others), beliefs about emotional expression, emotion suppression, positive and negative emotions and life satisfaction in general, controlling for depressive symptoms and neuroticism. It is assumed that the relationship between social anxiety and experiencing emotions and life satisfaction in general will be mediated by beliefs about expressing emotions and emotion suppression.
The further aim of the study was to test an interaction effect of social anxiety (with control of neuroticism and depression) and emotion suppression in explaining the frequency of experiencing positive and negative emotions.
The sample consisted of 521 female students attending University of Rijeka and University of Pula, in Croatia. The average age of participants was 21.21 (SD = 2.5 years; range 18-37).
To assess personality traits, The Big Five Inventory was used . It provides a good coverage of all five personality traits (Extraversion, Agreeableness, Conscientiousness, Neuroticism and Openness), and has satisfactory psychometric properties. Inventory consists of 44 items, using five-point Likert-type format for answers scoring. For the purposes of this study only Neuroticism subscale was used (8 items). Cronbach-alpha for the present sample was .81.
Beck Depression Inventory-II  has been used to assess depressive symptoms. It is a 21-item self-report scale, using four-point Likert-type format (higher number meaning more severe depressive symptom). Cronbach-alpha in this sample was .90.
Anxiety in social interaction was assessed using Social Interaction Anxiety Scale  and fear of being observed and evaluated by others using Social Phobia Scale . Both self-report scales consist of 20 items each, using five-point Likert-type format for scoring the answers. Cronbach-alpha for SIAS was .90 and for SPS .91.
Emotion Regulation Questionnaire  was used to assess emotional regulation strategies – reappraisal and suppression. For the purposes of this study only Suppression subscale was used. It consists of 4 items measuring the tendency to inhibit or conceal emotional expression that a person has experienced. Answers are scored by using seven-point Likert-type format. Internal reliability coefficient (Cronbach-alpha) for this subscale on the sample of participants of the present study was .74.
Attitudes Towards Emotional Expression Questionnaire  is constructed to measure negative beliefs and behaviors related to emotional expression and in the present study are used to assess beliefs about emotional expression. It is a 20-item self-report scale, using five-point Likert-type format for answers scoring. In the original form, the questionnaire consists of four factors: beliefs about meaning (sign of weakness), beliefs about expression (keep in control), beliefs about consequences (social rejection) and behavioral style (bottle up). The present study did not confirm these four subscales, but the authors recommended that subsequent research should focus on subscales as well as the overall scale. In the present study two factors were extracted, each of them with 10 items. The first factor is composed of items that are in the original questionnaire related to beliefs that expressing emotions is a sign of weakness, and beliefs that expressing emotions lead to social rejection. This factor is, therefore, called
To measure the subjective experience of emotion, Positive and Negative Affect Schedule – Expanded Form  has been used. This is a 20-item inventory that consists of 10 adjectives measuring positive affect (e.g. cheerful) and 10 adjectives measuring negative affect (e.g. irritable). Answers are scored by using five-point Likert-type format. Cronbach-alpha for positive affect subscale was .85 and for negative affect subscale .87.
In order to assess how a person is satisfied with her life, a Satisfaction with Life Scale  has been used. It is a 5-item self-report scale, using seven-point Likert-type format for answers scoring. Cronbach-alpha in the present sample was .86.
Data were collected during classes in group format, anonymously. Goal of the study was briefly explained and students participated voluntarily. The students who were not willing to participate were allowed to leave the room.
In order to determine the relationship between the variables involved in the study, correlation analyzes have been performed. Pearson's correlation coefficients are shown in Table 1.
|Anxiety in social interactions||.74**||.45**||.29**||.28**||-.30**||.52**||-.40**|
|Fear of being evaluated by others||.41**||.21**||.18**||-.21**||.53**||-.38**|
|Belief – expressing emotions leads to unpleasant consequences||.64**||.48**||-.15**||.39**||-.31**|
|Belief – emotions should not be expressed||.75**||-.14**||.14**||-.18**|
|Suppression of emotions||-.15**||.12**||-19**|
According to results, if a young woman has a higher anxiety in social interactions, she will have increased fear of other people's evaluation. Socially anxious person will believe that emotions should not be expressed and that their expression leads to unpleasant consequences so will suppress emotions and will experience negative emotions more often. Such a female experiences positive emotions also less frequently and is less satisfied with her life. The belief that expressing emotions leads to unpleasant consequences and that emotions should not be expressed are highly positively correlated with each other, and are positively correlated with suppression of emotions and more frequent experience of negative emotions. Both beliefs are negatively correlated with life satisfaction and positive emotions. A female who uses a strategy of suppressing emotions as a way of emotion regulation, has lower life satisfaction and less frequently experiences positive but more often negative emotions. Frequent experience of positive emotions means less frequent experience of negative emotions and greater satisfaction with life, while often experiencing negative emotions means less satisfaction with life in general.
In order to determine the unique relationship between social anxiety and other variables in further analyzes, a common variance of social anxiety shared with neuroticism and depression is controlled. The aim was to eliminate the possibility that the potential negative effects of social anxiety, primarily in experiencing positive emotions, can be attributed to a common variance, or negative affectivity, which is shared by social anxiety, neuroticism and depression, and not to the uniqueness of social anxiety. Certain models suggest that neuroticism as a higher common vulnerability factor explains most of the covariance among the more specific constructs such as social anxiety, depression and anger . It is considered that there are unique characteristics of high social anxiety that are not part of neuroticism. In order to control neuroticism and depression, regression analyzes were performed with standardized residuals calculated for both types of social fears. In regression analysis, the predictors included neuroticism and depression. Fear of social interaction was a criteria in the first analysis and a fear of being evaluated by others in the the second. After that the standardized values for both types of social fears were calculated. In this way we got two new variables, which were used for further analysis and in which negative affectivity related to neuroticism and depression is excluded, and only the part that is associated with a particular social fear has remained.
A variety of statistical analyzes were conducted in order to answer the research questions. The results were processed using the program LISREL 8  and SPSS 15.0 for Windows.
3.1. Testing the model
The tested model included the relationship between both types of social fears, beliefs about expressing emotions, suppression of emotions and frequency of experiencing positive and negative emotions and life satisfaction.
The theoretical model is shown in Figure 1. Only significant direct and indirect effects are shown.
Model fit indexes for this model are shown in Table 2.
The indexes shown in Table 1. indicate that this model is acceptable. Chi-square index is significant, but it is affected by sample size and for large samples is generally significant. To reduce sensitivity of model chi-square to sample size, ratio chi-square and degrees of freedom have been calculated. This ratio indicates that the model is acceptable as well as RMSEA value. The values of GFI, NFI and CFI show that the model has a good fit with empirical data.
In this model, the fear of other people's evaluation does not have direct or indirect effects on remaining variables included in the model. Anxiety in social interactions has only a direct positive effect on the belief that the expression of emotions leads to unpleasant consequences (.29), while other effects of this variable, except the one mentioned, are mediated by the belief that emotions should not be expressed and by suppression of emotions. The belief that expressing emotions leads to unpleasant consequences have a direct, positive and strong effect on the belief that emotions should not be expressed (.63) and moderate, negative and direct effect on positive emotions (-.38). The belief that emotions should not be expressed has a direct, high and positive effect on the suppression of emotions (.76), which has a direct negative effect on negative emotions (-.14). Positive emotions have a direct positive effect on negative emotions (.26) and negative on life satisfaction (-.14). Negative emotions have a direct negative effect on life satisfaction (-.16).
According to obtained results it is evident that the effects of social anxiety on the experience of positive and negative emotions and life satisfaction are achieved indirectly through beliefs about emotional expression and suppression.
3.2. The contribution of interaction effect of social anxiety and suppression of emotions to the frequency of experiencing positive and negative emotions
In order to test whether there is an interaction effect of social anxiety (with control of neuroticism and depression) and suppression of emotions in explaining the variance of experiencing positive and negative emotions after determining the individual contributions of both types of social anxiety individually and suppression of emotions, hierarchical regression analyzes were conducted.
Four hierarchical regression analyzes were conducted. As the first step, anxiety in social interactions has been included in the first two analyzes, and the fear of other people's evaluation in the other two. In each of the hierarchical regression analyzes the suppression of emotions has been included in the second step. In the third step the interaction of anxiety in social interactions and suppression has been included in the first two analyzes, and interaction of fear of other people’s evaluation and the suppression in the other two. For each analysis there were two criteria - positive and negative emotions.
Results are shown in Tables 3. and 4.
anxiety in social interactions
fear of being evaluated by others
suppression of emotions
suppression of emotions
anxiety in social interactions x suppression of emotions
fear of being evaluated by others x suppression of emotions
The results of hierarchical analysis which includes anxiety in social interactions show that included variables explain only 5% of the variance in frequency of experiencing positive emotions. Anxiety in social interactions explains 3% of the variance of criteria and suppression of emotions 2%, while the interaction of these two variables does not explain the frequency of experiencing positive emotions. Both anxiety in social interactions and suppression of emotions are negative predictors.
The results of hierarchical analysis which includes fear of being evaluated by others showed that only 3% of variance in frequency of experiencing positive emotions is explained, while suppression of emotions is the only significant and negative predictor.
anxiety in social interactions
fear of being evaluated by others
suppression of emotions
suppression of emotions
anxiety in social interactions x suppression of emotions
fear of being evaluated by others x suppression of emotions
The analysis with the anxiety someone is experiencing in social interactions, included in the first step, showed that only this variable is significant and positive predictor of the frequency of experiencing negative emotions and it explains 4% of the variance. Additional 1% of variance is explained by suppression of emotions as a strategy of emotional regulation, but as well as interaction included in the third step, it is not a significant predictor in the analysis.
The other conducted analysis showed that included variables explain only 5% of the variance in frequency of experiencing negative emotions. The fear of being evaluated by others and suppression of emotions are the only significant and positive predictors, while the interaction effect of these two variables is neither significant nor does it additionally explain the frequency of experiencing negative emotions.
When presented model for the presumed relations was tested, only social anxiety that refers to the fear that a person experiences during encounters with other people (e.g. initiating and maintaining conversations either in dyads or in groups) was found significant. As previous studies revealed its role in experiencing positive emotions, it was expected for this aspect of social anxiety to be significant [3, 10, 35, 36, 38, 40]. The second type of social anxiety, the one related to fear of being observed and evaluated by others, has not been studied enough in previous research. Although it was expected that, as a social fear, it could have had some effects on other variables, it was not proven. There is a possibility that this kind of fear is not so important in comparison with anxiety felt during social interactions.
Anxiety in social interactions did not show the expected direct effects. In the first place there was no direct effect on the beliefs that emotions should not be expressed and on their suppression. Its relationship with these variables is only indirect through belief that expression of emotions leads to unpleasant consequences with high and positive effects. This kind of relationship would mean that the suppression of emotions in a socially anxious woman occurs when she has a strong belief that expressing emotions leads to unpleasant consequences, due to which the belief that emotions should therefore not be expressed will be activated, which will result in her scruple of expressing any emotion (emotional suppression). The importance of our beliefs in the process of emotion regulation is confirmed by these results. Cognitive model  assumes that when an individual finds him/herself in a particular social context, negative assumptions concerning the assessment of the situation as dangerous will be triggered. In this way a whole series of negative automatic thoughts about themselves and other people are triggered. The finding that the beliefs about expressing emotions have an indirect role in relation of social anxiety and suppressing emotions stresses the significant role of cognitions. It seems that the belief that expressing emotions leads to unpleasant consequences is "superior" to the belief that emotions should not be expressed. This result is not unusual because if someone believes that it is not good to express emotions, he or she must have a direct or an indirect experience that emotion expression has led to unpleasant outcome. Unpleasant consequences are related to the belief that expressing emotions is a sign of weakness, which means that the one expressing emotions will be perceived as weak by others. Evaluating someone as a weak person means a specific trait or flaw because it is expected that an adult must be able to effectively and appropriately regulate his or her own emotional expression. Those who can control emotions and know how not to express them are estimated as powerful, while those who aren't successful in it tend to be estimated as weak, possibly even less desirable as friends and partners.
Although someone’s fear of being evaluated and perceived as weak is in the first place related to the aspect of emotion regulation, we might draw a parallel to research  which dealt with the fundamental fears that social phobics experience. One of the four dimensions of stimuli evaluation that elicit social fear refers to the fear of socially anxious individuals to be seen with some character flaws. Moscovitch  believes that this aspect can be represented with statements like "I'm boring", "I'm stupid“, with activities in which there is a disclosure of personal information (e.g. talking one on one) and with those in which these features are being questioned (e.g. telling jokes). According to his model, social phobics are afraid that their features, in fact themselves, in the eyes of others might be seen as incomplete in comparison to other people. It carries certain consequences of which they are afraid. Unpleasant consequences are related to other people's negative evaluation and rejection, embarrassment, loss of social status and likewise, which is neither pleasant nor desirable. For the same reason, an individual will resort to a number of safety behaviors. Being afraid of ranking as socially undesirable person supports the belief that expressing emotions leads to unpleasant consequences which clearly leads to belief that it is better to be emotionally restrained and, ultimately, to emotional suppression. Since socially anxious people simultaneously feel desire to approach other people but are afraid of rejection, strategy of suppressing emotional expression and adherence to the belief that expressing emotions leads to unpleasant consequences seems, from their perspective, a wise strategy to maintain social status. This mode of emotion regulation is used with intention to reduce the likelihood of experiencing a single unpleasant consequence. It is actually a paradox since the suppression of emotions is counterproductive, as it does not reduce unwanted, unpleasant experience and even strengthen physiological arousal [20- 22, 24, 26, 27, 51].
Anxiety in social interactions has only indirect effects on positive and negative emotions. The effect on positive emotions is realized over the belief that expressing emotions leads to the unpleasant consequences. Its effect on negative emotions is achieved through the emotional suppression (with previous indirect role of beliefs about expressing emotions). If a woman believes that expressing emotions is not good because it will lead to unpleasant consequences, she rarely experiences positive emotions. When she decides to suppress her emotions, it leads to increased incidence of experiencing negative emotions.
However, there are results that differ from the expected. It is found that the more frequent experience of positive emotions also leads to increased incidence of experiencing negative emotions and to overall life satisfaction reduction. This result might be due to low negative correlation (-.17) between experiencing negative and positive emotions, which was not expected when the emotions are measured in this way. When measuring the experiencing of emotions in a way that the participants are asked to indicate how they usually feel, there should be no correlation between positive and negative emotions . Increased emotionality of the participants (emotional reactivity) may be the reason for such result. Emotionality indicates individual's predominant intensity of emotional reactivity which includes a person's tendency to overreact even to weak stressors . Emotional reactivity was found to be associated with high blood pressure . People differ in emotional reactivity and those who are emotionally more reactive are lacking in control over thoughts related to the emotional content and therefore emotions themselves. It is possible that the participants in this study are prone to experience and to express positive emotions, which in turn has an effect on the more frequent experience of negative emotions. Generally speaking, it is possible that these women are more likely to experience positive and negative emotions. Some authors also suggest that people, who experience intense positive emotions, also experience intense negative emotions .
It is also uncommon to find that the emotional suppression leads to less often experience of negative emotions, which is not in accordance with paradoxical effect of this emotion regulation strategy [20-22; 24, 26, 27, 51]. It is possible that, despite the strong impact of the emotional expression beliefs, our participants do not come up with enough strong tendencies to inhibit or to conceal emotions (M = 3.05).
As expected, it has been confirmed that more frequent experience of negative emotions leads to reduced global life satisfaction . According to the hierarchical model of happiness , subjective well-being is related to cognitive and emotional components that are interconnected. Experiencing positive and negative emotions is emotional component and global assessment of life satisfaction refers to a cognitive component. According to
The obtained result that the more frequent experience of negative emotions leads to a reduced life satisfaction is in line with the assumptions on the assessment of quality of life, but the finding that more frequent experience of positive emotions has the same effect, certainly is not. As already noted, there are data on the intense experience of both positive and negative emotions  which can be related to the higher frequency of experiencing both of them. It turns out that the frequency of experiencing both positive and negative emotions have unfavorable effects on global life satisfaction for our participants.
It can be concluded that this model has provided additional insight into understanding the relationship between social anxiety, beliefs about the expression of emotions, suppression of emotions, and the experiencing of positive and negative emotions and the global life satisfaction in women. The main contribution of this model is the result of the role that cognitions play in these relationships. Their role is reveled over the beliefs about expressing emotions to other variables that confirmed the mediating role of these beliefs in relation to social anxiety and suppression of emotions . It has also been found that only aspect of social anxiety which refers to the anxiety experienced in social interactions has significant effects in these relations.
A further objective of this study was to examine whether there is an interactive effect of social anxiety (with control of neuroticism and depression) and suppression of emotions in explaining the frequency of experiencing positive and negative emotions after determining the individual contributions of each type of social anxiety and of emotion suppression. The starting point for setting this problem has been the assumption that severe social anxiety can become an even bigger problem if there are rigid tendencies in mastering and concealing emotional experiences . The authors were primarily concerned with the relationship of one aspect of social anxiety, the one that refers to the anxiety experienced in social interactions, and experiencing positive emotions, so this paper aims to determine the contribution of other type of social fear (fear of other people's evaluation) in experiencing positive and negative emotions. Although the research is primarily focused on understanding the experience of positive emotions in socially anxious women, contribution of both types of social fears and the suppression of emotions and their interaction effect on negative emotions has been tested.
The results of the hierarchical regression analyzes reveled that the interaction effect of social anxiety and suppression of emotion was not significant in explaining the frequency of experiencing neither positive nor negative emotions. This result does not support the theory about the interactional effects of these two variables on the expression of positive emotions , although the authors themselves have failed to confirm the „joint vulnerability“ model. The reason for this result may lie in the fact that women who participated in this study were low in social anxiety and suppression of emotion was not their main emotion regulation strategy (anxiety in social interactions M = 0.89, SD = 0.59; fear of other people's evaluation M = 0.73, SD = 0.61; suppression M = 3.05, SD = 1.22). It is possible that this result is a consequence of experiencing low frequency (or at least the reporting of it) of positive (M = 3.29, SD = 0.61) and negative emotions (M = 2.09, SD = 0.56), which again is the issue of emotional reactivity of the women involved in this research.
Only social fear related to anxiety experienced in social interactions was a significant predictor of the frequency of experiencing positive emotions. Expression of this type of fear contributes to less often experience of positive emotions. The results are consistent with findings about negative role of social anxiety in experiencing positive emotions (e.g. 3, 36). However, the role of anxiety in social interactions was confirmed, and the fear of other people's evaluation shows no significance in studying positive emotional experiences. Relationships with other people are very important for our welfare, and positive events and emotions are important for the development of such relations. Mastering fear of social interaction is important for the ability to develop relationships with other people and, to some extent, precedes fear of the other people’s evaluation which could emerge after the contact. In this study, fear of other people’s evaluation mostly includes evaluation of foreigners and people who are not emotionally important to us (which does not diminish the importance of this type of evaluation for socially anxious people). Women in this study are not diagnosed as socially anxious, so it is possible that this fear does not interfere with their relationships with other people as far as the fear of interaction with them do. The presence of other people is important for the frequency of experiencing positive emotional experiences, and if this fear is prevalent, the opportunity for such experiences is reduced.
Suppressing was found as a negative predictor of the frequency of experiencing positive emotions, which means that the use of this strategy of emotion regulation contributes to less often experience of positive emotions. The result is consistent with the finding that the suppression of emotions leads to decreased expression of both positive and negative emotions, thus interfering with relationships with other people  and that this strategy is associated with rare experiencing positive emotions and their seldom expression [21, 24, 27]. Suppression of emotions was significant predictor of experiencing negative emotions, but only in the analysis which included fear of other people's evaluation as a predictor. Selecting this emotion regulation strategy means more frequent experience of negative emotions. The result is consistent with previous studies that have found this mode of emotion regulation as counterproductive because it leads to paradoxical reinforcement of physiological arousal and unwanted affect itself , and the suppression of expressing negative emotions does not bring any relief in terms of the subjective experience of negative emotions .
Women who have expressed any of the two types of social fears experience more frequent negative emotions which is consistent with previous results. Anxiety disorders are associated with exaggerated and persistent negative emotions  and the relationship between social anxiety and frequent experience of negative emotions is confirmed in a number of studies (e.g. 3).
Results of this study showed that only anxiety in social interactions explains the experience of positive emotions with only 3% of the variance explained. In his meta-analysis Kashdan  also found that social anxiety explains 4-5% of the variance in positive experiences after controlling depression. Our study went a step further by controlling neuroticism as a personality trait, which would certainly „blur” independent contribution of social anxiety in explaining the experience of positive emotions. The data of this study provide significant contribution, showing that social anxiety has its own independent role in understanding the reduced experience of positive emotions, and that it cannot be attributed to effects of depression and neuroticism.
It was found that both types of social fears have a significant role for experiencing negative emotions and that each could explain 4% of the variance of frequency of experiencing negative emotions. This finding confirms and emphasizes the independent role of social anxiety in more frequent experiencing of negative emotions because effects of depression and neuroticism are controlled.
The results of this study should be considered within the context of its limitations. First, the study is based on participants' self-assessment. While this is the most common method of data collection, for this type of research it is important to use a clinical sample of socially anxious people who use different strategies of emotion regulation. In order to comprehend better this set of problem, experimental design would have been a better solution to answer the research questions. However, the results of previous studies have shown different ways of regulating emotion in the laboratory experiments and those implemented in everyday circumstances, so it would be better to implement this type of research in everyday circumstances of socially anxious individuals. Such research would better succeed to grasp impairment of social functioning in relation to emotional regulation strategies, as well as effects on close relationships.
Further limitation is the fact that only women participated in the study. Because of gender differences in the severity of social anxiety and ways of regulating emotions, future research should check the same model in a male sample and compare models for both sexes.
The sample included only those participants who agreed to participate in the study and who, on average, were not socially anxious. It is possible that some women who decided not to participate in the research were more socially anxious and their results would be valuable in the study of relationships of social anxiety and other variables. Research on a clinical sample of socially anxious women whose daily functioning is disrupted by disorder could be especially useful and might give different results.
Future research should focus on examining individual differences in emotional reactivity and sensitivity when studying the relationships that are examined here. As this study deals only with one of the strategies of emotion regulation - suppression, it would be important to test the model with reappraisal as emotion regulation strategy. This strategy has been found as adaptive and, as our data indicate the importance of cognitions in relationship between social anxiety and suppression of emotions, it would be important to see their relations when using this cognitive strategy of emotion regulation.
Considering that dimensions of social anxiety and emotion regulation are important for creating our relationships with others, future research might include assessment of quality of close relationships in the model. It was found that emotion regulation strategies have different effects on memory, and through the memory contents that are related to interpersonal relationships on the quality of close relationships [24, 60].
Practical contributions of this research are also worth mentioning. The results point out the importance that beliefs about emotions and their expression have on suppression of emotion, and experiencing emotions in general. Since these are dysfunctional beliefs that the expression of emotions leads to unpleasant consequences (to which a person does not want to be exposed), and due to which the belief that emotions should not be expressed is activated, the therapeutic work should focus on restructuring such beliefs about danger of emotions and their expression. On the other hand, the expression of emotions is important in interpersonal relationships. In development of quality close relationships it is necessary to mutually share emotional experiences which does not occur if someone perceives it as threatening. Therefore, it is important to teach socially anxious people about adaptive strategies of emotion regulation and to point out benefits, advantages and disadvantages of using different strategies. However, as the emotion regulation strategy is only a part of the entire system of self-regulation, it would be useful to check the person’s capacities for self-regulation in general.
In conclusion, the results of this study have provided more insight into the complex relationship between social anxiety, emotional experiences and the quality of life in general. The mediation mechanisms that play a role in these relationships have been revealed by structural modeling, which has not been done in previous researches.