Running Head: VALUING HAPPINESS AND DEPRESSION Desperately seeking happiness: Valuing happiness is associated with symptoms and diagnosis of depression

Culture shapes the emotions people feel and want to feel. In Western cultures, happiness is an emotion that many people want to feel. Although experiencing happiness is associated with increased well-being and psychological health, recent evidence suggests wanting to feel happy to an extreme degree, or, highly valuing happiness leads to decreased well-being. To examine whether these effects of valuing happiness might extend to clinical outcomes, we examined the hypothesis that depression is associated with highly valuing happiness. To do so, we examined the relationship between valuing happiness and depression in two U.S. samples. As hypothesized, valuing happiness was associated with increased depressive symptoms in a community sample with remitted major depressive disorder (MDD), even when controlling for social desirability and neuroticism (Study 1). Furthermore, valuing happiness was elevated in a remitted MDD sample (vs. healthy controls), even when controlling for current depressive symptoms, general affect valuation, and extreme goal pursuit (Study 2). Taken together, these findings suggest that the culturally-pervasive value placed on attaining happiness can represent a risk factor for symptoms and a diagnosis of depression. More broadly, they indicate that a cultural approach can meaningfully extend our understanding of clinical phenomena. Keywords: Depression, emotion, valuing happiness VALUING HAPPINESS AND DEPRESSION 3 Desperately seeking happiness: Valuing happiness is associated with symptoms and diagnosis of depression Feeling happy is one of the most highly endorsed values in Western cultures (Ford & Mauss, in press). This is not surprising: Westerners are often exposed to messages and artifacts promoting the pursuit of happiness (e.g., the Declaration of Independence). In spite of this, there is also substantial variation in how much any given person wants to pursue happiness across cultures (Tsai, Knutson, & Fung, 2006) and within cultures (Eid & Diener, 2001). Examining these individual differences within cultures is important for several reasons (Kitayama & Park, 2007). First, understanding the effects and correlates of values within a culture is an important part of understanding the effects and correlates of values between cultures and how values within cultures interact with values between cultures (e.g., Eid & Diener, 2001). Further, not all people from one cultural context are the same. Within-culture variation can elucidate the effects of culturally-shaped factors while holding constant other factors that plague between-culture research (e.g., language, exposure to Western culture; e.g., Cohen, 2009; Snibbe & Markus, 2005). Finally, within-culture variation in values also has important implications for psychological health (Ford & Mauss, in press). Examining these individual differences therefore presents an opportunity to better understand the factors that shape psychological health. The goal of the present research was to add to this literature by examining the links between valuing happiness and psychological health within the U.S. cultural context. The Paradoxical Effects of Valuing Happiness Many people in the U.S. strongly value happiness (Ford & Mauss, in press), and American culture appears to have almost an obsession with happiness (e.g., Gruber, Mauss & VALUING HAPPINESS AND DEPRESSION 4 Tamir, 2011). And to clarify terminology, although happiness is sometimes examined as a broader construct incorporating well-being and psychological health (Diener, Suh, Lucas, & Smith, 1999), consistent with prior work, we use this term to refer to a positive affective state (e.g., Russell, 2003). Even children’s storybooks produced in the U.S. are more likely to portray very happy characters, compared to storybooks from other cultures (e.g., Taiwan; Tsai, Louie, Chen, & Uchida, 2007). The ideas promoted by these cultural artifacts (e.g., storybooks) causally reinforce individual’s personal values for happiness (e.g., young children who read these storybooks are more likely to want to engage in activities that promote greater levels of happiness) (Tsai, Louie, et al. 2007). Recent findings suggest a surprising paradoxical effect of these culturally shaped ideas – valuing happiness can actually make its attainment less likely. Several lines of evidence support the hypothesis that highly valuing happiness is associated with lower well-being, decreased happiness, increased depressive symptoms, and compromised social outcomes (Mauss et al., 2012; Mauss, Tamir, Anderson, & Savino, 2011; Schooler, Ariely, & Loewenstein, 2003; Tamir & Ford, 2012a). Importantly, several experimental studies support the causal relationship between valuing happiness and maladaptive outcomes. For example, participants experimentally induced to value happiness were less happy after a happiness emotion induction compared to those not induced to value happiness (Mauss et al., 2011). It has yet to be demonstrated, however, whether these negative outcomes translate to the clinical domain, including severe and chronic psychiatric disorders like major depression. Valuing Happiness and Depression Major depressive disorder (MDD) is a chronic, pervasive, and highly impairing psychiatric disorder (Murray & Lopez, 1997). It is thus critical to identify mechanisms that may help explain the etiology and recurrence of MDD. One such mechanism may be the values VALUING HAPPINESS AND DEPRESSION 5 people hold regarding their emotions, like happiness. At least two lines of evidence support the notion that valuing happiness may increase risk for the onset and maintenance of depression. First, valuing happiness decreases positive emotion (Mauss et al., 2011), and decreased positive emotion is one of the hallmark features of depression both conceptually (American Psychiatric Association, 2000) and empirically (Bylsma, Morris, & Rottenberg, 2008). Second, valuing an emotion is a critical early stage in the emotion regulatory process (Tamir & Mauss, 2011) because holding this value makes subsequent emotion regulation attempts more likely (Tamir & Ford, 2012b). Importantly, if these values become extreme or inflexible (e.g., believing one must feel happy all the time), they may contribute to disordered emotion regulation. Disordered emotion regulation, in turn, is a key component of depression (Gross & Muñoz, 1995) and has been implicated in the onset of depression (Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008). In sum, there are multiple ways in which highly valuing happiness may be linked with – and perhaps lead to – depression. Yet, no study to date has examined the clinical implications of valuing happiness. To address this empirical gap, we conducted two studies examining the link between highly valuing happiness and depression within diverse samples in the U.S. Study 1 Study 1 assessed the relationship between valuing happiness and depressive symptoms in a sample of participants currently remitted from recurrent MDD. Participants were in remission but had a history of three or more MDD episodes in their lifetime, and were thus highly likely to experience a future MDD episode (Judd, 1997). We also assessed and controlled for socially desirable responding to further ensure the link between valuing happiness and depressive symptoms is specific to highly valuing happiness rather than susceptibility to social demand. VALUING HAPPINESS AND DEPRESSION 6 Methods Participants. Participants were a community sample (N=98) recruited from the Denver metropolitan area. The Structured Clinical Interview for DSM-IV (SCID-CT; First, Williams, Spitzer & Gibbon, 2007) confirmed that participants were currently remitted from MDD and had a history of at least three prior episodes of MDD (M=13.4 episodes, SD=22.4). Participants with comorbid Axis I or II disorders were excluded to isolate the links between happiness valuation and depression specifically. Demographic and clinical characteristics are provided in Table 1. Materials. Valuing happiness. The valuing happiness scale (Mauss et al., 2011) consists of seven items measuring to what extent participants highly value happiness (e.g., “Feeling happy is extremely important to me”) rated on a scale of 1 (strongly disagree) to 7 (strongly agree) that were averaged to create a mean valuing happiness score (α=.79). Depressive symptoms. The Beck Depression Inventory (BDI; Beck, Steer, & Brown, 1996) consists of 22 items assessing current depressive symptoms rated on a scale of 0 (e.g., I do not feel sad) to 3 (e.g., I am so sad or unhappy that I cannot stand it) that were summed to create a composite depressive symptom score (α=.86). Social desirability. The Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991) consists of 40 items assessing to what extent participants agree with socially desirable statements (e.g., I never regret my decisions) on a scale of 1 (not true) to 7 (very true) that were averaged to create a total social desirability score (α=.77). Neuroticism. The neuroticism subscale of the International Personality Item Pool (Goldberg, 2005) consists of 10 items assessing negative emotionality and reactivity (e.g., I worry about things) on a scale of 1 (very inaccurate) to 5 (very accurate) that were averaged to create a total neuroticism score (α=.83). VALUING HAPPINESS AND DEPRESSION 7 Procedure. Participants were recruited as part of a randomized controlled trial investigating the effect of psychosocial interventions on preventing depression relapse. All relevant study measures were collected before randomization to treatment had occurred. After obtaining informed consent, the SCID-CT was administered by trained interviewers. Several days later (M=9.3 days, SD=7.6, range=0-48), participants reported their demographics, valuing happiness, depression symptoms, neuroticism, and social desirability in an online survey. Results and Discussion Valuing happiness was correlated with increased depressive symptoms, r=.27, p=.008, even when using partial correlations to control for social desirability, r=.22, p=.029, neuroticism, r=.20, p=.045, and demographic characteristics, r=.34, p=.004 (including gender, age, race (“White” vs. other), relationship status (partnered vs. not), years of education, and annual family income). Thus, valuing happiness was associated with greater depressive symptoms in people diagnosed with a recurrent history of MDD, a sample at particular risk for developing depression. This sample was socioeconomically diverse and recruited from the community, thus supporting the external validity and generalizability of our findings. However, it is unclear whether the relationship between valuing happiness and depression holds for self-reported depressive symptoms only, or if this relationship generalizes to clinician-rated symptoms and to a clinician-assessed diagnosis of MDD. Study 2 Study 2 measured valuing happiness and clinician-rated symptoms of depression in a sample diagnosed with remitted MDD and a healthy control sample. Study 2 expands on Study 1 in four critical ways. First, by examining the correlation between valuing happiness and clinician-rated symptoms of depression, we avoid the limitations of relying only on self-reported VALUING HAPPINESS AND DEPRESSION 8 data. Second, by comparing levels of valuing happiness for those with remitted MDD to a control group, we can ascertain whether higher valuing of happiness is associated with a clinical diagnosis of MDD, in addition to elevated depression symptoms. Third, by controlling for clinician-rated current depressive symptoms when comparing valuing happiness of those with remitted MDD to the control group, we can examine whether there is a trait-like association between valuing happiness and a history of depression. This would suggest that valuing happiness might be a risk factor for depression, and not simply a symptom of depression. Finally, by assessing and controlling for the extent to which participants (a) generally value negative and positive emotion and (b) generally pursue goals to an extreme degree, we were able to test whether the association between valuing happiness and depression is specific to the extreme and inflexible valuing of happiness, or if the association is accounted for by (a) valuing positive (or negative) emotion broadly speaking or by (b) generally pursuing goals in the extreme. Methods Participants. The Structured Clinical Interview for DSM-IV (SCID-IV; First, Spitzer, Gibbon, & Williams, 1997) confirmed that participants were either currently remitted from MDD (n=31) or healthy controls (CTL) who did not meet current or past criteria for any DSM-IV-TR Axis I disorder (n=30). Participants were recruited using online advertisements and flyers posted in New Haven, CT and surrounding communities. For the MDD group, the average age at onset was 16.05 years (SD=7.36), the average illness duration was 14.92 years (SD=10.33), and the lifetime average of major depressive episodes was 5.60 (SD=7.44). Demographic and clinical characteristics are provided in Table 1. Materials. Valuing happiness. Participants completed the same valuing happiness scale as in Study 1 (α=.70). VALUING HAPPINESS AND DEPRESSION 9 Clinician-rated depressive symptoms. The Inventory of Depressive Symptomatology (IDS-C; Rush, Gullion, Basco, Jarrett, & Trivedi 1996) is a 30-item, clinician-rated measure of depressive symptoms experienced in the past week. Current remitted status for the MDD group was verified according to both SCID-IV criteria and cutoff scores on the IDS-C (≤11). General valuing of negative and positive emotion. The Affect Valuation Index (AVI) (Tsai et al., 2006) measures the extent to which participants “ideally” want to experience nine positive emotions (e.g., excited, happy, relaxed or “positive affect valuation”; α=.81) and nine negative emotions (e.g., lonely, fearful, hostile or “negative affect valuation”; α=.73). Extreme goal pursuit. The Drive subscale of the Behavioral Approach System/Behavioral Inhibition System Scale (Carver & White, 1994) consists of 4 items measuring the extent to which participants engage in the persistent pursuit of desired goals (“I go out of my way to get things I want”) (α=.80). Procedure. Participants were recruited for a study on “mood and emotion.” After obtaining informed consent, the SCID and IDS-C were administered by trained interviewers. Participants then reported their demographics and completed a series of procedures not relevant to the present study. At the end of the session, participants completed the valuing happiness scale, the AVI, and the Drive subscale. Results and Discussion Valuing happiness was correlated with current depressive symptoms, r=.28, p=.029, even when using partial correlations to control for positive and negative affect valuation, r=.34, p=.01, extreme goal pursuit, r=.28, p=.029, and demographic features, r=.31, p=.032 (including gender, age, years of education, annual household income, race (“Caucasian” vs. other), VALUING HAPPINESS AND DEPRESSION 10 relationship status (partnered vs. not) and employment (employed vs. other)). See Table 2 for correlations between valuing happiness and control variables. We next examined whether the MDD and CTL groups differed in valuing happiness. A univariate ANOVA revealed that the MDD group reported higher levels of valuing happiness (M=3.97, SD=1.01) than the CTL group (M=3.18, SD=0.91), F(1, 60)=10.48, p=.002, ηp=.15. This finding also held when controlling for all demographic factors, F(1, 47) =5.59, p=.022, ηp=.11. Further, we were able to address several alternative hypotheses by controlling for various confounding variables as covariates in the above univariate analysis. First, we verified that the group difference remained significant when controlling for current depressive symptoms, F(1, 58)=5.88, p=.018, ηp=.09, suggesting that valuing happiness is a characteristic of MDD, and does not simply contribute to the negative mood symptoms associated with the disorder. Second, we verified that the group difference remained significant when controlling for the general valuing of positive and negative affect, F(1, 58)=8.55, p=.005, ηp=.13, suggesting that the link between valuing happiness and depression is not simply due to depression leading people to value the emotions they lack (i.e., happiness). If this were the case, depressive symptoms should be associated with increased valuing of happiness, whether that valuing is extreme or not. Our results confirm that this is not the case: depressive symptoms were not significantly associated with the increased general valuing of positive emotion and the link between valuing happiness and depression remained significant when controlling for the general valuing of positive emotions in a partial correlation and ANCOVA. Thus, it is not the case that depression is simply associated with wanting to feel missing emotions. This further suggests that VALUING HAPPINESS AND DEPRESSION 11 valuing happiness to an extreme degree is a risk factor for depression, and not necessarily the other way around. Third, we verified that the group difference remained significant when controlling for extreme goal pursuit, F(1, 59)=9.92, p=.003, ηp=.14, suggesting that the link between valuing happiness and depression is not simply due to extreme goal pursuit in general. It is possible that pursuing any goal in the extreme could be associated with maladaptive outcomes like depression and thus, extreme goal pursuit may drive the maladaptive nature of highly valuing happiness. We can begin to rule out this hypothesis and suggest that specifically valuing happiness – and not generally pursuing goals – is a risk factor for depression. In sum, these results indicate that valuing happiness is associated with clinician-rated depressive symptoms and that valuing happiness is stronger for those with remitted MDD (vs. control participants), even when controlling for clinician-rated depressive symptoms, general valuing of positive and negative emotions, and extreme goal pursuit. This points to valuing happiness as a trait-like risk factor for a diagnosis and maintenance of MDD. General Discussion Across two studies, we demonstrated that highly valuing happiness is associated with symptoms and a history of clinical depression. This research extends prior work on the psychological-health correlates of valuing happiness in several ways. First, the association between valuing happiness and depressive symptoms was observed in clinical samples, whether those symptoms were self-reported (Study 1) or clinician-rated (Study 2). This relationship also held when controlling for social desirability, neuroticism, extreme goal pursuit, and several potentially confounding demographic features (gender, age, education, income, race, relationship status, and employment). Furthermore, there is evidence to suggest that it is the extreme nature VALUING HAPPINESS AND DEPRESSION 12 of valuing happiness, rather than valuing happiness in general, that is associated with depressive outcomes (given that the link holds when controlling for general affect valuation). The external validity of these findings is strengthened by the socioeconomic and clinical diversity of our samples in Study 1 and 2. Although studying depression often involves measuring how people currently feel, the present results suggest that the emotions people want to feel may also play an important role in depression. Valuing Happiness and Depression The present results suggest that highly valuing happiness is associated with elevated symptoms and diagnosis of MDD. We propose that valuing happiness acts as a risk factor (and not only a correlate or an outcome) for four reasons. First, experimental manipulations that induce strong valuing of happiness causally contribute to increased negative emotion during positive mood inductions (Mauss et al., 2011; 2012). Although these studies were conducted in non-clinical samples, they support the notion that valuing happiness can cause negative outcomes. Second, research on acceptance-based therapies suggests that interventions targeting emotional values can improve depressive outcomes (Ma & Teasdale, 2004). Specifically, accepting one’s emotions, and not inflexibly striving for one specific kind of emotion such as happiness, appears to causally contribute to greater psychological health. Third, within the present investigation, higher valuing of happiness was unlikely to be a consequence of the current experience of depressive symptoms (i.e., people who currently feel bad want to feel better) for two reasons: the relationship between valuing happiness and depression held independent from current symptoms and depression was not associated with the general valuing of positive emotion (thus reaffirming that higher valuing of happiness was not due to depression VALUING HAPPINESS AND DEPRESSION 13 leading people to value the emotions they lack, like happiness). In sum, these findings support that highly valuing happiness acts as a risk factor for depression. What mechanisms could account for effects of valuing happiness on depression? First, experimental work has shown that valuing happiness decreases positive emotion (Mauss et al., 2011), and positive emotion is known to be decreased in depression (Bylsma et al., 2008). Second, extreme and inflexible emotional values may lead to disordered emotion regulation (Tamir & Mauss, 2011), and disordered emotion regulation is integrally involved in depression (Gross & Muñoz, 1995). Although we outline several arguments in support of the notion that valuing happiness causally contributes to the genesis and maintenance of depression, and although at least two plausible mechanisms exist for this link, future experimental and longitudinal studies are needed to confidently draw causal conclusions about valuing happiness. Valuing happiness and psychological health are likely related in a bi-directional manner, and future research involving repeated assessments of valuing happiness and depression will help to untangle what is likely to be a complex interplay. Culture, Valuing Happiness, and Depression The present research suggests that culturally shaped emotion values are related to serious psychological health outcomes. Specifically, we propose that the extreme valuing of happiness is a component of depression in Western contexts. This research contributes to a growing body of research examining the links between culturally shaped emotion values and psychological health (e.g., Tsai et al., 2006). As such, the present research lies at the intersection of cultural and clinical approaches to psychological health. Given the pervasive belief that happiness is good, linking the extreme valuing of happiness to depressive outcomes is novel and meaningful. Importantly, within and between culture variability in the extent to which people value happiness VALUING HAPPINESS AND DEPRESSION 14 (Eid & Diener, 2001) leaves room for this to be a critical individual difference that can influence psychological health. In the current investigation we examined valuing happiness as a within-country factor. By examining this question within the U.S., we were able to isolate the relationship between valuing happiness and depression while avoiding confounds with other factors (e.g., language, exposure to Western culture). Examining variation within one country is a stepping-stone toward understanding between-culture variation as well as about the interaction of betweenand withinculture variation. Examining within-culture variation, however, is not a substitute for cross-cultural examinations, and measuring valuing happiness and its correlates across cultures is an important future direction. For example, there is some reason to believe that happiness may be a universal value. However, happiness can have different meanings across cultures (Lu & Gilmour, 2004; Uchida, Norasakkunkit, & Kitayama, 2004). Thus, cultures may still vary in how much happiness is valued and in how strongly psychological health outcomes (e.g., depression) are associated with this value. Indeed, such values may be further utilized to examine cultural differences in risk factors for depression (Chentsova-Dutton et al., 2007) or manifestations of depression (Ryder et al., 2008). Limitations and Future Directions We highlight three additional limitations and directions for future research. First, the present research focused on depression. Examining other types of mood disorders would advance our understanding of whether valuing happiness is a shared risk factor for the development and maintenance of multiple mood disorders. Second, the effect sizes were fairly modest, though consistent across two studies. Even modest effect sizes, however, can exert critical cumulative VALUING HAPPINESS AND DEPRESSION 15 influences on psychological health. Finally, not everyone who values happiness develops depression. There may be key boundary conditions under which the pursuit of happiness is not associated with negative outcomes (e.g., Ford & Mauss, in press; Layous & Lyubomirsky, in press). Concluding Comment While happiness is strongly valued in many contexts, as a culturally shaped emotion value, there is also much variation in how strongly people endorse the extreme valuing of happiness. The present research suggests that highly valuing happiness is a key correlate of – and potential risk factor for – depression. These results highlight the importance of examining culturally shaped values to gain a more complete picture of psychological health. VALUING HAPPINESS AND DEPRESSION 16