Fears of Negative Emotions in Relation to Fears of Happiness, Compassion, Alexithymia and Psychopathology in a Depressed Population: A Preliminary Study

Objectives: While fears of negative or aversive emotions are linked to experiential avoidance and psychopathology, recent studies have also focused on the relation between psychopathology and fear of positive emotions. This study explores 1. which negative emotions of anger, anxiety and sadness on most feared and avoided and 2. the links between fears and avoidance of negative emotions, with fears of positive and affiliative emotions, alexithymia, and self-reported depression anxiety and stress. Method: A new scale was developed to measure fears of three negative emotions anxiety anger and sadness. 52 participants suffering from moderate to severe depression completed this measure, along with fear of happiness, fears of compassion, alexithymia and psychopathology. Results: Interestingly fears of negative emotions were not correlated with each other; in other words one can be frightened of one negative emotion but not another. The correlation between the fear of an emotion and the avoidance of that emotion was different for the three negative emotions, with fear of anger being the most strongly linked to its avoidance. Fear of sadness was the only feared ‘negative’ emotion associated with depression. Fear of sadness and fear of anger, but not anxiety also linked to fears of positive emotions and alexithymia. Conclusions: Fears of (so called) negative emotions vary in terms of the degree to which people are fearful of them and avoid them. Importantly it was sadness, a neglected emotion in the studies of emotion avoidance, which accounted for the higher proportion of variance for depression and alexithymia.


Introduction
Experiential avoidance, linked to fear of emotion, is a central premise of Acceptance Commitment Therapy [1][2][3]. There is increasing evidence that fear of 'so called' negative or aversive emotions are significantly associated with psychological difficulties [4][5][6][7][8][9]. We are using the term 'negative emotions' here for ease of exposition but of course many of these emotions have positive functions, are important for healthy functioning and the avoidance of danger and harm. So correctly speaking they are threat emotions. We also recognize that there are a number of measures for different dimensions of fear of emotions [10,11] but these are not specifically focused on threat or negative emotions..
Looking at specific emotions then, fear of becoming anxious is associated with panic disorder, agoraphobia and other anxiety disorders [12]. Fear of embarrassment and negative evaluation are core fears in social phobia leading to the avoidance of situations that could lead to unfavorable evaluations. Fear of sadness has been recognized in a range of disorders [5]. Avoidance of sadness has been associated with complicated grief [12], and recent evidence suggests that clients with borderline personality disorder struggle with tolerating sadness [13].
Fear of anger, and suppression of anger, are related to the relative social power and rank in the relationship in which it occurs [14], that is to say people are more likely to express anger to those less powerful themselves than more powerful themselves [15]. So person may express anger to their children but not to their boss. Some depressed people are more expressive of anger while others are indeed fearful and suppress their anger [16]. Anger expression and suppression have been central to many psychodynamic views of depression and other disorders for nearly hundred years [16]. Brody et al. [17] found that recovered depressed outpatients were more frightened and more suppressing of anger than controls. However, Gilbert et al. [18] found this was partly linked to gender. In both men and women depression was linked to internalized hostility, but only in men was depression also a linked externalizing hostility. Women were more likely to internalize blame, be submissive and suppress anger than men. Gilbert et al., [19] found that ruminating on anger is also linked to depression.

Positive emotions
An increasing number of studies has also found that some people can have a fear of 'positive emotions', such as feeling safeness, happiness, and affiliative emotions associated with compassion [9,[20][21][22]. Arieti and Bemporad [23] suggested that some depressed patients have what they called 'a taboo on pleasure' and are fearful of positive emotions. They may have beliefs such as "happiness never lasts -when I feel happy I am always waiting for something bad to happen". Positive emotions can be associated with negative outcomes particularly in children who have been punished at times when they have been enjoying themselves [24]. In a recent study in a student population, Gilbert et al. [21] found that fear of happiness is strongly correlated with depression. Beblo et al. [9] also found that fear and suppression (inhibiting the outward signs of inner feelings) of positive and negative emotions were linked to depressive symptoms. The authors highlight the importance of not only encouraging patients to participate in enjoyable situations but actually practicing allowing one to feel pleasant emotions. Raes et al. [25] found that dampening positive affect predicts depressive symptoms and is also related to symptoms of panic disorder, social phobia, generalized anxiety disorder and OCD [26]. Weeks et al. [27][28][29] found that some socially phobic people are fearful of positive evaluations as well as negative evaluation.
People can fear positive emotions linked to activation and arousing feelings of pleasure and joyfulness [21,22]. In addition positive emotions linked to affiliative emotions of closeness, intimacy, feeling loved and being cared for and about can also be feared [21,30,31]. Compassion is linked to these affiliative emotions and recent research suggest that some individuals are frightened of receiving and/or giving compassion Fear of compassion has been found to be linked to fear of happiness [21,30,31] but currently it is unknown the degree to which the fear of affiliative emotion is linked to fears of negative emotions.
Another core theme linked to the fear of positive emotions is a need for greater clarity on the type positive emotion along the dimensions of hedonic and eudaimonic [32]. Hedonic happiness is based on the experience of pleasure and the pursuit of pleasure however it is arrives. Eudaimanic pleasure in contrast is linked to pursuing virtues, acquiring skills and wisdom and leading a "meaningful life". For those individuals pursuing the latter and particularly for those with a spiritual quest, life hedonic pleasures by themselves can be avoided as somewhat trivial and frivolous [33]. So the experience of positive emotion does values as would be suggested by contextual behaviour therapy.

Difficulty in processing any emotion
Another dimension related to difficulties in processing emotions is called alexithymia [34]. This refers to a set of interconnected difficulties of identifying and distinguishing between feelings and the bodily sensations of emotional arousal; difficulty describing feelings (especially to other people) and a stimulus-bound, externally oriented cognitive style with constricted imaginal processes, as evidenced by a paucity of fantasies [35]. Hence, alexithymic individuals typically think the causes of their feelings are external rather than internal.
Taylor et al. [36] suggest that alexithymia constitutes a risk factor not only for psychiatric and behavioural problems but also for medical and physical health problems, particularly when these are influenced by disordered affect regulation. In a recent review of alexithymia and chronic diseases, Baiardini et al. [37] note that alexithymia has been linked to many somatic and organic (particularly immune-mediated) diseases and also to unhealthy behaviours. The authors argue that it is possibly a failure to successfully regulate negative affect (due to an inability to adequately process emotions) that leads to negative coping behaviour in alexithymics. The sources of alexithymia are many and although the fear of emotions may be one factor contributing to alexithymia it is unknown if there are specific feared emotions and the degree to which avoidance of emotional experiencing is linked to alexithymia.

Aims and hypothesis
This study set out to explore the link between the fear of a negative emotion and the avoidance of that same emotion, the links between them, and their association with fear of positive emotions (happiness and compassion). We also explored how these fears of emotions are associated with the psychopathology measures of depression, anxiety and stress. We are aware of various scales that measure different types of difficulty in the experiencing of emotion. For example, Williams et al. [11] developed a scale that measures a generalized fear of emotion, which focuses on a fear of loss of control over one's emotional experience. Leahy's [10] emotional schema questionnaire measures a number of attitudes about emotions such as shame, incomprehensibility, being overwhelmed. However, we wanted to develop a simple measure of 'fear of the emotion' and the extent of its avoidance. It is not inevitable that a feared emotion would be avoided and so require separate assessment. We also hypothesised that the link between avoidance and the fear of emotion may not be the same for each negative emotion. We expected to find significant links between fears of negative emotions and fears of positive emotions. Also we hypothesize that different negative emotions may link to depression, anxiety and stress in different ways.

Method
Participants 52 participants were recruited from Community Mental Health Teams (CMHTs, N=40) and from self-help groups (N=12) within Derbyshire. They were 36 women and 16 men with an age range of 21 to 70 years (M =48.38; SD=13.75). All participants were currently experiencing moderate (N=2), severe (N=8) or extremely severe (42) depression (based on the score on the depression subscale of the DASS scale, Lovibond & Lovibond) [38].
All participants completed self-report scales measuring psychopathology, fears of happiness, fears of compassion and fears of negative emotions and alexithymia.

The Fear of Negative Emotions Scale
This newly developed scale contains six items which explore people's fear and avoidance of three negative emotions of: Anger, anxiety and sadness. For each emotion there were two questions. The first asked how frightened the person is of the feelings of an emotion. The second question asked how much they went out of their way to avoid the feelings of that same emotion (e.g. for anxiety the two questions are "I'm frightened of the feelings of anxiety" and "I go out of my way to avoid feeling anxious"). Respondents are asked to rate how characteristic each of the statements is for them on a five-point Likert scale scaling from 0 ("Not at all like me") to 4 ("Extremely like me"). Items were generated from statements made during therapy sessions with author PG and from the literature (e.g., "I am frightened of the feelings of anxiety").

Fear of Happiness Scale
This 9-item scale explores people's perceptions and anxieties around feeling happy and having positive feelings in general (e.g. "I feel I don't deserve to be happy"). Items concern the extent to which each statement is true to the participants and are rated on a five-point Likert scale ranging from 0 ("Not at all like me") to 4 ("Extremely like me"). Cronbach's alpha for this scale was .90 [21].

Fears of Compassion Scales
These three scales measure fear of compassion for self (compassion we have for ourselves when we make mistakes or things go wrong in our lives), fear of compassion from others (the compassion that we experience from others and flowing into the self), and fear of compassion for others (the compassion we feel for others, related to our sensitivity to other people's thoughts and feelings).
In this study we used the 'Fear of receiving compassion from others' (e.g. "Feelings of kindness from others are somehow frightening" -13 items) and the 'Fear of compassion for self' (e.g. "I feel I don't deserve to be kind and forgiving to myself -15 items) scales. We did not use the 'Fear of giving compassion to others' scale as the aim of this study was to explore fears of experiencing one's own positive and negative emotions and difficulties identifying one's emotions. Respondents rated on a five-point Likert scale how much they agree with each statement (0="Don't agree at all" to 4="Completely agree"). The Cronbach's alphas for these scales were .80 for fears of receiving compassion from others and .83 for fears in giving compassion to self [30].

Toronto Alexithymia Scale (TAS-20)
This 20-item scale measures three factors of alexithymia: 1) difficulty identifying feelings, which assesses the ability to identify and distinguish feelings (e.g. "I have feelings that I can't quite identify"); 2) difficulty describing feelings, measuring the ability to describe feelings to others (e.g. "It is difficult for me to find the right words for my feelings"); and 3) externally oriented thinking, which considers the tendency for individuals to focus their attention externally (e.g. "I prefer to just let things happen than to understand why they turned out that way"). The items are rated on a five-point Likert scale ranging from "I Strongly disagree" to "I Strongly agree". Cronbach's alphas for these scales are .80 for Factor 1; .76 for Factor 2; .71 for Factor 3; and . 86 for the total scale [40]. The TAS-20 uses cutoff scoring: equal to or less than 51=non-alexithymia, equal to or greater than 61=alexithymia. Scores of 52 to 60=possible alexithymia.

Data analysis
Analysis was conducted using SPSS version 19 for PCs. The data were checked for normality of distribution and outliers using scatterplots which revealed no outliers. Means and standard deviations are shown in Table 1. To explore differences between participants we used t-tests. To address our primary research objectives, the analysis focused on correlations and multiple regression to explore the relationship between the variables in this study.

Descriptive analysis
The skewness values ranged from -0.136 to -.946 and kurtosis values ranged from -1.2 to 3.71. The means, standard deviations and Cronbach's alphas of the variables studied are shown in Table 1. The internal consistency of the scales was generally good, with the exception of the alexythimia subscales which had Cronbach's alphas between .53 and .68.

Mean
Std. Deviation

Correlation analysis
Pearson's Correlation Coefficients (two-tailed) for items on the fear and avoidance of negative emotions scale are presented in Table 2. Correlations between fears and avoidance of negative emotions, and fear of happiness, fears of compassion from others and for self, and depression anxiety and stress are presented in Table 3.

Fears and avoidance of different negative emotions
The first two columns in Table 2 (in light grey shade) show the correlations between fear of anger, fear of anxiety and fear of sadness. The last two columns of Table 2, (in darker grey shade) show the correlations between the avoidance of anxiety, avoidance of sadness and avoidance of anger. It's notable that the correlations between fears of negative emotions are not significant, indicating that in this depressed population fear of one negative emotion does not imply fear of another. Avoidance items are significantly correlated with a particularly higher correlation between avoidance of anger and avoidance of anxiety. Table 2 are the correlations between fear of each negative emotion and its corresponding avoidance. Interestingly, fear of anxiety is significantly but minimally correlated with avoidance of anxiety, whereas for anger there is a high correlation between the fear of anger and the avoidance of anger. The fear of sadness is also relatively highly correlated with its avoidance. So the link between fear of an emotion and avoidance of an emotion is not straightforward but dependent on the type of emotion.    Table 3 provides the correlations between fears and avoidance of negative emotions with depression, anxiety and stress. It is notable that fear of feeling sadness and to a lesser degree avoidance of sadness come out as the only fears linked to depression with fears and avoidance behaviours of the other emotions not significantly associated. For anxiety, only the fear of feelings of anxiety is significantly correlated. For stress, only avoidance of anger is significantly correlated.

Fears and avoidance of negative emotions and fears of positive emotions
The second half of Table 3 provides data on the link between fear and avoidance of negative emotions and fear of happiness, fear of compassion towards self and from others. Fear of happiness is significantly correlated with fear of sadness, fear of anxiety and fear of anger. Fears of compassion are significantly linked to fear of sadness and fear of anger but not fear of anxiety.

Fears and avoidance of negative emotions and alexithymia
We also explored how the fears and avoidance of negative emotions link to alexithymic traits. Alexithymic traits are linked to the fear of sadness and anger but not anxiety. Alexithymia is linked to the avoidance of all three negative emotions.

Discussion
There is increasing interest in the nature of the fears of emotions both positive and negative, and the degree to which individuals react to these fears with avoidance. For this study we developed a short fear and avoidance of negative emotions scale focusing on three specific emotions of anger, anxiety and sadness. The data is interesting in a number of ways. First, in this small depressed population we found that being afraid of one negative emotion is not necessarily correlated with being afraid of another one. In contrast avoidance of different emotions appeared to be linked, especially in the case of avoidance of anxiety and avoidance of anger. This indicates again that fear of an emotion and avoidance needs to be considered separately. Indeed, as ACT maintains, many people may well engage with feared emotions if they are linked to core goals or values (e.g., facing anxiety to pass a driving test).
We also found that the fear of different negative emotions had different relationships with avoidance. While the association between the fear of anxiety and the efforts to avoid anxiety seemed to be low, the association between fear of sadness and avoidance of sadness and particularly the association between fear of anger and avoidance of anger was found to be much stronger. This suggests that it is important to be specific about the kinds of emotions people are frightened of and avoid. This also points to important individual differences which require attention in therapy. So it seems that those who fear sadness may not fear anger and vice versa. This is important because the fear of anger has long been linked to depression [16] and recently fear of sadness has been indicated to be an important, avoided emotion for people with borderline difficulties Lecours and Bouchard [13]. Interestingly, in this study too, the fear of sadness turned out to be the only feared negative emotion linked to depression.
We were also interested in the link between the fears of negative emotions and fears of happiness, and compassion from others and to self. Fears and avoidance of sadness and anger are positively linked with fearing happiness and fearing compassion from others and to self. The link between fear of sadness and fear of positive emotions found in this study fits with suggestions by Gilbert and Irons [41] that grief and sadness can be 'frozen' in some individuals and that this inability to process loss and sadness may have a general suppressant effect on emotion processing. Indeed, individual patients undergoing compassion focused therapy [42] have revealed that going through a grieving process for the attachment traumas and for the childhoods they wanted but didn't have, facilitated a recovery of a range of feelings Gilbert and Irons, [41]. One patient of author PG explained it was 'sort of learning to see in colour'. In regards to alexithymia, fear of sadness was the most significant feared negative emotion.
Emotional avoidance is central to a number of therapies especially ACT [24] and Emotion Focused Therapy [2]. Taken together, this study indicates that fear and avoidance of emotions might be more strongly linked depending on the emotion. Fear and avoidance of sadness may be potentially more pathogenic than often recognised, having been found to be the only fear of a negative emotion linked with depression. It also seems to play an important role in problematic emotional processing as measured by alexithymia. Importantly, sadness in linked to grief. Changes that take place under guided grief process may have the function of healing and enabling integration of avoided feelings and memories [41]. Indeed, Gilbert & Irons pointed to the lack of research on personalised grief (grief for the life one wanted, or a childhood lost to abuse, in contrast to grief for losses of others; e/g, early death of one mother) and the importance of facilitating that process as part of the therapeutic journey [41][42][43][44][45].

Limitations
There are a number of weaknesses to the study. This is a small study with relatively simple measures. Second this is not a study of diagnosed depressed patients using objective interview data. Rather it is a preliminary study on the link between depressed symptoms in a heterogeneous depressed population and fears of different types of emotion. Therefore we cannot say if these relations between negative emotion and depression will pertain in more tightly defined diagnosed populations, or indeed if they vary between different types of clinical difficulty. In addition as a cross-sectional study no prediction can be made about causality. The scale we developed has face validity with just one item measuring each construct but it cannot be validated against other scales since none exist that measure these domains in this way.
Future studies may also wish to look agenda variation. For example is depression related to feel sadness in both men and women or are they different patterns of fears of negative emotions. Larger numbers are required for this analysis. Here we will simply interest in developing a simple scale and offering preliminary data in terms of its potential value in terms of avoidance of negative emotion.