Everyone wants to be happy. Indeed, happiness may be the ultimate fundamental ‘goal’ that people pursue in their lives (Diener, 2000), a pursuit enshrined as an inalienable right in the US Declaration of Independence. The question of what produces happiness and well-being is the subject of a great deal of contemporary research, much of it falling under the rubric of ‘positive psychology’, an emerging field that also considers issues such as what makes for optimal relationships, optimal group functioning, and optimal communities. In this article, we first review some prominent definitions, theories, and research findings in the well-being literature. We then focus in particular on the question of whether it is possible to become lastingly happier in one's life, drawing from our recent model of sustainable happiness. Finally, we discuss some recent experimental data suggesting that it is indeed possible to boost one's happiness level, and to sustain that newfound level.
A number of possible definitions of happiness exist. Let us start with the three proposed by Ed Diener in his landmark Psychological Bulletin (1984) article. The first is ‘leading a virtuous life’, in which the person adheres to society's vision of morality and proper conduct. This definition makes no reference to the person's feelings or emotions, instead apparently making the implicit assumption that reasonably positive feelings will ensue if the person toes the line. A second definition of happiness involves a cognitive evaluation of life as a whole. Are you content, overall, or would you do things differently given the opportunity? This reflects a person-centered view of happiness, and necessarily taps peoples’ subjective judgments of whether they are satisfied with their lives. A third definition refers to typical moods. Are you typically in a positive mood (i.e., inspired, pleased, excited) or a negative mood (i.e., anxious, upset, depressed)? In this person-centered view, it is the balance of positive to negative mood that matters (Bradburn, 1969). Although many other conceptions of well-being exist (Lyubomirsky & Lepper, 1999; Ryan & Frederick, 1997; Ryff & Singer, 1996), ratings of life satisfaction and judgments of the frequency of positive and negative affect have received the majority of the research attention, illustrating the dominance of the second and third (person-centered) definitions of happiness in the research literature.
Notably, positive affect, negative affect, and life satisfaction are presumed to be somewhat distinct. Thus, although life satisfaction typically correlates positively with positive affect and negatively with negative affect, and positive affect typically correlates negatively with negative affect, these correlations are not necessarily strong (and they also vary depending on whether one assesses a particular time or context, or the person's experience as a whole). The generally modest correlations among the three variables means that an individual high in one indicator is not necessarily high (or low) in any other indicator. For example, a person with many positive moods might also experience many negative moods, and a person with predominantly good moods may or may not be satisfied with his or her life. As a case in point, a college student who has many friends and rewarding social interactions may be experiencing frequent pleasant affect, but, if he doubts that college is the right choice for him, he will be discontent with life. In contrast, a person experiencing many negative moods might nevertheless be satisfied with her life, if she finds her life meaningful or is suffering for a good cause. For example, a frazzled new mother may feel that all her most cherished life goals are being realized, yet she is experiencing a great deal of negative emotions on a daily basis. Still, the three quantities typically go together to an extent such that a comprehensive and reliable subjective well-being (SWB) indicator can be computed by summing positive affect and life satisfaction and subtracting negative affect.
Can we trust people's self-reports of happiness (or unhappiness)? Actually, we must: It would make little sense to claim that a person is happy if he or she does not acknowledge being happy. Still, it is possible to corroborate self-reports of well-being with reports from the respondents’ friends and family. When this is done, reasonably good agreement is usually found between self-reports and friend-reports, indicating that self-reports do correspond to something observable in the real world (Diener, 1994; Sandvik, Diener, & Seidlitz, 1993).
A cynic might ask: Does happiness even matter? Or, is the attempt to understand and promote happiness just another symptom of decadent Western culture's bourgeois self-obsession? In fact, happiness matters a great deal. A recent and comprehensive meta-analysis revealed a wide variety of benefits that accrue from positive emotion and well-being, including greater career success, better relationship functioning, increased creativity, enhanced physical health, and even longer life expectancy (Lyubomirsky, King, & Diener, 2005). This meta-analysis amassed considerable longitudinal data supporting the proposition that happiness leads to success (rather than vice versa), with mean effect sizes of .21 for the happiness, satisfying relationship link; .24 for the happiness, satisfying work link; and .18 for the happiness, health/longevity link. Thus, happiness is worth promoting, not only because it feels good, but because it is a wise social and public health investment.
In considering the research on happiness, we begin by examining some basic demographic and life-circumstantial variables that are associated with greater SWB. An obvious place to start is income. Are wealthier people happier? The answer is ‘yes, but not as much as you'd think’. In one meta-analysis of 85 studies, the correlation between income and SWB was only .17 (Haring, Stock, & Okun, 1984). Furthermore, this association typically has a curvilinear component, such that variations in income make the most difference at low levels of income; beyond a certain point of basic sufficiency, income has a smaller effect (Argyle, 1999; Diener & Biswas-Diener, 2002; Diener & Diener, 1995). Indeed, very well-off individuals are only slighter happier than the blue-collar workers they employ (Diener, Horwitz, & Emmons, 1985). Thus, those who are convinced that they would be much happier if they could only make $200,000 instead of $100,000 appear to be in error!
A somewhat larger effect of wealth has been observed at a between-country level of analysis, with citizens of richer countries typically describing themselves as considerably happier than citizens of poorer countries (Diener, Diener, & Diener, 1995). However, even this effect seems to be driven by a lack of basic self-sufficiency (i.e., unmet basic needs) in the poorest nations; among nations with adequate average income, national differences in SWB are driven by other factors besides gross domestic product, such as democratic government and support for equal rights (Diener et al., 1995). Another intriguing fact is that as countries become wealthier over time, their citizens’ happiness does not increase (Diener & Oishi, 2000; Lane, 2000). Although many Americans have reached a level of affluence that would have amazed us 50 years ago (Easterbrook, 2003), happiness levels have not budged over those 50 years. This suggests that instead of focusing on enhancing gross domestic product, governments should instead focus on enhancing gross domestic happiness! And indeed, this is now the primary public policy goal of the Kingdom of Bhutan, a small country in the Himalayan mountains.
The conclusion of the research discussed so far then is that money is a necessary, but not a sufficient, determinant of happiness. What about other demographic predictors of SWB? Argyle (1999) concluded that chronological age has a small positive correlation with SWB, at least for some SWB measures (e.g., Charles, Reynolds, & Gatz, 2001; Diener & Suh, 1998). Although the relationship between age and happiness is complex and does not always emerge, both cross-sectional and longitudinal studies have found that older people tend to be happier than their younger peers. For example, a 22-year study of healthy veterans revealed that well-being increased over the course of these men's lives, peaked at age 65, and did not start significantly declining until age 75 (Mroczek & Spiro, 2005).
Education has an even smaller correlation with well-being than does age, which is mostly accounted for by the occupational status (as well as income) associated with greater education (Witter, Okun, Stock, & Haring, 1984). As this suggests, social status and class also have (small) effects on SWB. A slightly larger effect is found for marriage, with married people (especially women) being happier than unmarried people (Inglehart, 1990), although this effect is due in part to happier people being more likely to attract marriage partners (Lyubomirsky, King, & Diener, 2005). Religious people are also somewhat happier, although again, the effect is rather minor (cf. Ellison & Levin, 1998).
As this list illustrates, a surprising discovery of the research on demographic correlates of SWB is that overall such effects are small. Indeed, these life-circumstantial factors may account for less than 10% of the variance in happiness (Andrews & Withey, 1976), although Diener (1984) suggested that the figure may be as high as 15%. This leads naturally to the question: What accounts for the rest? It appears that biological, psychological, and behavioral variables must by some means determine the majority of happiness.
Perhaps the single most important determinant of SWB is genetics (Lykken & Tellegen, 1996; Tellegen et al., 1988). Simply put, some people arrive in this world with a predisposition to cheerfulness, optimism, and joy, whereas others are born with a predilection toward fearfulness, pessimism, and depression. Studies of twins separated at birth have yielded heritability estimates for SWB ranging from .40 to .70, with the most common figure around .50. This means that if you had been cloned at the time of your birth (or had had an identical twin, unbeknownst to you), with the clone (or twin) being raised in a different part of the world, your SWB today could be well predicted by the clone or twin's SWB! In personality trait terms, peoples’ characteristic levels of neuroticism and extraversion (the traits most closely related to SWB) are already in place at birth, and will remain largely stable across the lifespan. These personality traits appear to be rooted, ultimately, in peoples’ basic settings on core features of biological temperament, including approach, avoidance, activation, and inhibition (Buss & Plomin, 1984).
In short, an unsettling implication of the heritability data is that SWB cannot be changed. You and your clone will, of course, vary over time, sometimes falling below and sometimes rising above the SWB level predicted by your genes (although the particular genes involved are yet undiscovered). However, you will both tend to keep returning to the same baseline level – a ‘regression to the mean’ effect that is strong enough to generate the large correlation listed above – despite the fact that your within-person variations would likely be out of sync with one another. If SWB is indeed unchangeable, several implications follow: (i) The ‘pursuit of happiness’ guaranteed by the US Declaration of Independence is folly, (ii) positive psychology's attempts to promote and enhance peoples’ SWB are futile, (iii) self-help books and the people who buy them are deluded, and (iv) we might as well get used to and resign ourselves to the basic happiness level we are dealt. Since ‘what goes up must come down’, trying to become happier may be doomed to failure, perhaps even leading to unnecessary unhappiness!
A related reason why happiness increase attempts may be futile is hedonic adaptation (see Frederick & Loewenstein, 1999, for a review). People are very good at adjusting to their circumstances, so that what was originally a source of joy (or irritation) becomes part of the unnoticed background, losing its power to impact them. Indeed, the well-known study of Brickman, Coates, and Janoff-Bulman (1978) suggested that even recent lottery winners and paralysis victims tend to return to their original happiness baselines, after experiencing initial boosts or decrements, respectively. Gilbert, Pinel, Wilson, Blumberg, and Wheatley (1998) have referred to a ‘psychological immune system’ that protects people from suffering too long or maladaptively from negative events; unfortunately, this immune system may also prevent us from experiencing lasting enhancements in SWB. As this reasoning suggests, SWB may operate like a homeostatic system; each individual may have a ‘set point’ for SWB to which she keeps returning, in the same way that her body resists departures from weight and other metabolic set points.
Are long-term changes in SWB impossible, then? Not necessarily. First, recall that the heritability of SWB is only .50, a large number to be sure, but one that still allows considerable room to maneuver compared to other inherited biological features (such as eye color). Second, as previously mentioned, chronological age is positively (although modestly) correlated with SWB (Argyle, 1999). This finding in itself indicates that SWB can increase over a period of decades, rather than remaining permanently stable. Third, emerging longitudinal research suggests that despite the strong heritability of personality traits, people do tend to move towards more positive trait profiles as they age, in particular becoming lower in neuroticism as they approach middle age (Wood & Roberts, 2006). Research by Sheldon and colleagues suggests that one source of these positive age-related changes is older peoples’ ability to resist social pressures and to pursue goals for more self-endorsed reasons, the result of a normative maturational process (Sheldon & Kasser, 2001; see also Carstensen, Isaacowitz, & Charles, 1999). Other reasons for optimism include the fact that psychotherapy can have a measurable and lasting positive impact upon peoples’ mood and adjustment (Smith, Glass, & Miller, 1980), the fact that some well-being interventions have shown a degree of success (Fava, Rafanelli, Cazzaro, Conti, & Grandi, 1998; Fava et al., 2005; Fordyce, 1977, 1983), and the fact that starting a happy marriage has been shown to have lasting positive impacts on SWB for some individuals (Lucas, Clark, Georgellis, & Diener, 2003). Further consistent with the notion that well-being can be permanently altered, Lucas (2007) summarized research showing that events such as divorce, death of a spouse, unemployment, and disability are associated with lasting negative changes in SWB. Although in one sense this is distressing news, in another sense it supports our argument that the set point is not destiny.
Figure 1 presents a pie chart (reprinted from Lyubomirsky, Sheldon, & Schkade, 2005) summarizing what we have observed so far: that about 10% of the variance in SWB can be explained by relatively static demographic and circumstantial factors and that about 50% of the variance can be explained by genetics (i.e., unchanging personality, and temperament). We (Lyubomirsky, Sheldon, & Schkade, 2005; Sheldon & Lyubomirsky, 2004) have argued that the remaining 40% of the variance must perforce by determined by what people do– that is, the activities with which people fill up their days, with greater or lesser degrees of success and enjoyment. Our model of sustainable happiness thus focuses in particular upon the various forms and qualities of intentional activity that people engage in, as the crucial determinant of variation in SWB. The model also specifies that people can sustainably increase their SWB, if they appropriately and intentionally tailor their lives.
To illustrate, consider this within-subject regression equation:
- SWBt=β0(set point) +β1(effects of circumstances) + β2(effects of activity) + error
According to this equation, we can predict a person's SWB at a particular time t if we know about her genetics and fixed temperament (i.e., her set point), her current SWB-relevant demographic/circumstantial profile (i.e., income, health, marital status, religious affiliation), and her current activities (i.e., what goals and practices she is pursuing, for what reasons, and with how much success). The equation specifies that the genetic set point is the default or intercept value, which drives the outcome when the effects of all the other variables are equal to 0. However, circumstances and activities can detract from or enhance this default value, depending on a number of factors (to be considered below).
The SWB equation suggests that the set point may be better conceptualized as a set range in which an individual can vary. In other words, each person may have a characteristic range or ‘attractor space’ in which he or she might be found at a given time, with larger deviations from the center of that range being increasingly difficult and unlikely. Consider an analogy: Genetics do not determine a person's exact adult IQ score (i.e., 105); instead, they establish a range of potential (i.e., 92 to 118), with the person's exact location in that range being determined by developmental factors and both stable and transient environmental influences. SWB may be considered in a similar way (although, of course, IQ does not fluctuate the way SWB does). In short, although some people may never reach a state of delirious joy, they can still attain (or not attain) their own highest happiness potentials. From this perspective, the happiness-increase question becomes: How can one reach, and stay in, the upper end of one's set range?
It is worth noting, again, that a person's life circumstances do affect his happiness, but these effects tend to be short lived and small because of hedonic adaptation to static features of his life. Therefore, Lyubomirsky, Sheldon, & Schkade (2005) argued that intentional activity has the best potential to elevate people into the upper end of their happiness range, and even to keep them there –if certain conditions are met. The key is to achieve and maintain a state in which one has many positive experiences over time, and that these experiences (i) fit one's personality, dispositions, and needs; (ii) vary in their content so that the effects of hedonic adaptation are minimized; and (iii) vary in their timing, again to minimize the effects of hedonic adaptation. For example, if an individual adopts the intentional activity of running regularly in order to stay in shape, then it is important that (i) running in and of itself is at least somewhat interesting and enjoyable (as compared to other physical activities he could select, such as playing tennis or doing yoga), (ii) he runs in a mix of places and in varying ways so that the resulting experiences stay ‘fresh’ and novel, and (iii) he runs at appropriate times (i.e., when excited about exercising or when feeling a strong need to exercise because of recent inactivity). To use another example, an academic researcher would be well advised to make a habit of looking for new questions and projects to tackle, ideally ones that fit her evolving interests and ideas. She should then pursue these projects in new and varying ways (i.e., learning novel skills and techniques along the way), so that interest and discovery are maximized and the feeling of rote routine is minimized. In the process, a steady stream of positive experiences will enter the researcher's consciousness, providing a route to enhanced SWB over and above the effects of her set point.
Does this mean that one should strive for complete spontaneity, never making plans and always ‘going with the flow?’ Not at all. Self-discipline and routine do not detract from SWB – in fact, they can contribute to its enhancement. Lyubomirsky, Sheldon, & Schkade (2005) distinguished between two types of habit: (i) habitually initiating an activity, and (ii) habitually performing the activity the same way every time. The first habit is likely to be good for SWB, as it helps people to ‘get over the hump’ to undertake activities with happiness-boosting potential (as long as they are not too rigid about the activity's timing). However, the latter habit is likely to undermine SWB, as it tends to make the activity boring and rote, sticking the person in a rut and depriving him or her of the fresh and novel experiences necessary to avoid adaptation. For example, if the runner mentioned above wishes to be happier, it would be adaptive for him to develop the habit of running regularly, but it would not be adaptive for him to run the exact same route every time. Similarly, a person trying to cultivate an ‘attitude of gratitude’ should make a habit of pausing to count his or her blessings, but should not always dwell on the same blessings; new sources of gratitude should be identified and appreciated.
As these examples illustrate, our model is quite consistent with ‘bottom-up’ theories of SWB, which argue that it is the cumulative sum of small experiences that matters (Diener, 1984), because people judge their happiness by consulting (i.e., integrating over) memories of their lives. The more positive and novel the recent experiences one can recall, the higher one will rate one's happiness; in contrast, positive but taken-for-granted experiences do not contribute as much to the judgment, and recalled negative experiences not surprisingly detract from it. As one example of a bottom-up research approach, Sheldon and Elliot (1999) showed that the semester-long accumulation of small satisfying experiences in undergraduates (involving feeling autonomous, competent, and related in one's daily activities) predicted enhanced global SWB at the end of the semester. Simply put, the more positive and meaningful experiences one has along the way, the greater one's ultimate judgments of well-being.
However, the sustainable happiness model also accommodates top-down theories of SWB, which posit that one's stable attitudes, beliefs, and personal stands can impact SWB by influencing the overall subjective construal of experience (Diener, 1984). Resolving to have a positive attitude (i.e., ‘Always look on the bright side of life’) is similar to developing a stable habit to initiate a certain activity (‘Find time to meditate at least 3 times a week’). In either case, stable cognitive structures are instantiated that can have steady top-down influence on the quality of subsequent experience.
As this reasoning illustrates, the sustainable happiness model construes ‘activity’ very broadly, potentially encompassing cognitive as well as behavioral forms of activity. The model also encompasses volitional activity, as observed when people pursue particular personal goals. For example, Sheldon and Elliot (1999) showed that, ultimately, it was the successful pursuit of self-appropriate goals that produced a greater accumulation of small satisfying experiences in daily life, which in turn predicted enhanced global SWB. In other words, adopting the new volitional activity of pursuing a defined set of goals (a stable, top-down factor) led to a better quality of daily experience (a varying, bottom-up factor), which led ultimately to enhanced SWB.
In addition to the longitudinal goal research cited above, considerable empirical evidence is emerging to specifically support the various facets of the sustainable happiness model. For example, Sheldon and Lyubomirsky (2006a) conducted three studies to compare the effects of circumstantial change and activity change upon SWB. In a preliminary investigation (study 1), they asked participants to sign up for the study only if ‘there has been some significant positive change in the circumstances of your life since the beginning of the semester, which has given you a boost since it occurred. Circumstances are facts about your life, such as living arrangement, monetary situation, or course load. For example, you may have moved to a better dorm or better roommate, received an increase in financial support so you can have more fun, or dropped a course that you were really going to have trouble with.’ At the same time, another study description read: ‘Please sign up only if you have adopted some significant positive new goal or activity since the beginning of the semester, which has given you a boost since it occurred. Goal/activity means something you chose to do or get involved in, which takes effort on your part. For example, you may have joined a rewarding new group, club, or sports team, decided on a major or career direction which makes it clear what to focus on, or taken on some other important new project or goal in your life.’
After selecting into one of the two conditions, participants rated their current SWB. To assess the effects of hedonic adaptation and variety, respectively, they also responded to two questions: ‘To what extent have you gotten used to the change, that is, to what extent do you find that you've become accustomed to it, such that it doesn't give the same boost as before?’ and ‘To what extent is the change something that varies over time, i.e., something that adds variety to your life?’ The activity-change group reported higher positive affect than the circumstance-change group, an effect that was partially accounted for by the fact that they had not yet gotten ‘used to’ their change, and the fact that their change was providing more variety in their lives.
In a subsequent longitudinal study (study 2), participants’ SWB was measured three times – at an initial time point and twice more in the ensuing weeks. At time 2, participants rated the extent to which they had experienced activity and circumstantial changes (the study's predictors) since time 1. Analyses revealed considerable stability in SWB over the three time periods, demonstrating the influence of the set point. However, there was also some variation in SWB. In fact, whereas both activity and circumstantial changes were associated with enhanced SWB at Time 2, only activity change was still associated with enhanced SWB at Time 3. Thus, although participants who experienced positive circumstantial changes experienced a boost in SWB, this boost was transient and short lived. In contrast, the boost experienced from positive activity changes was more lasting, presumably because of the varied positive experiences it provided and because hedonic adaptation was less likely to occur. This pattern was replicated in a second longitudinal study (study 3), using better measures of circumstance and activity changes and additional measures of SWB.
In summary, the article of Sheldon and Lyubomirsky (2006a) provided evidence for a key distinction within the sustainable happiness model, between activity and circumstantial change. Still, the research was limited by the fact that participants self-selected into the two change conditions. Research in progress is demonstrating that randomly assigning participants to make an activity or circumstantial change can produce a similar pattern, if participants succeed in making the change they promised they would (Sheldon & Lyubomirsky, 2007).
Notably, other recent research has begun to support the basic postulates of the sustainable happiness model, specifically those concerning critical moderators of the effects of activity assignments. First, it turns out that participants actually have to do the activity in order to obtain benefits; significant effects are only observed for those who follow through with the assigned activity (Seligman, Steen, Park, & Peterson, 2005; Sheldon & Lyubomirsky, 2006b). This may seem a truism, but it is an important issue to consider when researchers ask participants to do something for their own good, just as when physicians prescribe treatments to their patients (DiMatteo, 2004).
With respect to the ‘optimal timing’ moderator proposed by the sustainable happiness model, Lyubomirsky, Sheldon, & Schkade (2005) reported data from two 6-week-long experiments in which participants were asked to either ‘count five blessings during the week’ or to ‘commit five acts of kindness during the week’. Treatment participants’ changes in SWB were later compared to those of nonintervention controls. The ‘acts of kindness’ experiment included an additional factor – namely, ‘spread the five acts out over the week’ or ‘commit all five acts in the same day’. Committing all five acts on the same day produced a measurable increase in longitudinal SWB, whereas spreading them out did not. This finding suggests that it can be costly to have to always keep the new activity in mind, and not so beneficial if the new positive experiences merely trickle in as a result. However, if one commits to really focusing on the activity on just one day a week, one may minimize the cognitive costs while deriving the strongest experiential benefits – benefits that are likely to be remembered later when global SWB judgments are being made (however, see Diener, Sandvik, & Pavot, 1991, for evidence that the frequency of positive affect is more important than the intensity of positive affect for predicting happiness).
Providing more evidence for an optimal timing moderator, Lyubomirsky, Sheldon, & Schkade (2005) asked participants to think about five things for which they were grateful (i.e., a healthy body, my parents) either once a week or three times a week. Relative to controls, participants who expressed gratitude indicated greater SWB 6 weeks later, but only those who expressed gratitude just once a week; participants instructed to express gratitude three times a week did not benefit from the exercise. Once again it appears that assignments that require participants to do something too often or too regularly, so that it has to be brought to mind more often, may backfire compared to assignments in which participants are asked to do the activity on fewer occasions. During the initial adoption phase of a new positive activity, when that activity is most vulnerable to becoming extinct, it may be best to provide oneself with strong jolts of positive experience, thereby reinforcing the desire to keep doing the activity, but not to provide those jolts too frequently, so that the activity does not lose its freshness and meaning and, hence, turn into a chore.
In addition to considering the timing of an activity, the sustainable happiness model also postulates that varying the activity itself is important, because this mitigates the effects of hedonic adaptation. Supporting this proposed moderating factor, Tkach (2005) demonstrated that participants who were randomly assigned to vary the types of kind acts they would perform on a weekly basis showed higher levels of happiness and well-being 10 weeks later relative to those who did not vary their kind acts, and relative to comparison controls. Similarly, Sheldon and Lyubomirsky (2007) showed that the rated variety of the experiences deriving from activity and circumstance changes mediated the differential effects of these two changes on SWB.
Yet another postulate of the sustainable happiness model is that the activity should fit or match the person's personality, disposition, and needs. Such ‘self-concordance’ is important because it can provide the activity with stable sources of motivational energy from which to draw (Sheldon, Ryan, Deci, & Kasser, 2004). In other words, if the new activity aligns with chronic motives, values, interests, and dispositions within one's personality, it is likely to be continually funded within the psychic economy (Sheldon & Elliot, 1998). Sheldon and Lyubomirsky (2006b) showed just such an effect, as participants who reported self-concordant motivation (i.e., more interest and identification, and less pressure and guilt) for counting their blessings or for contemplating their best possible selves were more likely to perform, and thus benefit from, their assigned activities. Similarly, Dickerhoof, Lyubomirsky, and Sheldon (2007) showed that self-concordance predicted greater benefit from practicing optimism and expressing gratitude. These newer experimental findings are broadly consistent with past personal goals research showing the many SWB-related benefits of self-concordant goal pursuit (Sheldon & Elliot, 1999; Sheldon & Kasser, 1998). Together, the findings suggest that self-concordant motivation, which we construe as one indicator of person-activity fit, is important no matter what the activity.
Dickerhoof et al. (2007) also investigated a moderating factor that was not raised in the original sustainable happiness model (Lyubomirsky, Sheldon, & Schkade, 2005) – whether participants self-selected into a ‘happiness-enhancement study’ or instead self-selected into a neutral-sounding (i.e., ‘cognitive exercises’) study. This is an extremely important issue for happiness researchers to address: If a treatment only works when people want and expect it to work, this would severely limit the conclusions we could draw concerning the treatment's efficacy. To illustrate the problem, consider a recent study that found that people who engaged in one of 5-week-long exercises (such as daily generating a list of three things that went well or using their personal strengths in novel ways each day) experienced increases in well-being over the short and long term (somewhat depending on the exercise; Seligman et al., 2005). Notably, all of the participants in this study were self-selected, being presumably highly invested in their participation and knowing full well that it was supposed to make them happier. The same is true, of course, of the purchasers of self-help books, enrollees in metaphysical programs, and consumers of alternative therapies. Perhaps peoples’ claims to benefit from these activities ultimately reflect only dissonance reduction efforts, placebo effects, or demand characteristics? Does it even matter what participants are assigned to do, as long as they think it will make them happier?
Indeed, we found that happiness self-selection was important in order for the assigned activities to work (Dickerhoof et al., 2007). Those who self-selected into the ‘cognitive exercises’ study did not experience SWB benefits from doing the exercises, despite being told upon arriving at the lab that the exercises consisted of positive practices that were expected to make them happier. (The two groups differed only in the initial means of enrolling in the study; once at the laboratory, the cover story and procedures were identical.)
Although, on the face of it, this finding suggests that happiness intervention effects are merely placebo effects, Dickerhoof et al. (2007) also found evidence for a significant interaction that speaks against the placebo interpretation. Specifically, the self-selection factor only produced greater SWB in the two positive activity conditions (cultivating gratitude and counting blessings), and not in the neutral control condition in which participants merely wrote about the details of their day. Thus, it appears that both motivation to pursue happiness and an appropriate happiness pursuit strategy are needed in order to effect change. Initial positive expectancies and beliefs are not enough.
We conclude by summarizing what we have discovered so far about happiness enhancement. First, the pursuit of happiness takes work (i.e., the practice of intentional activity), but hopefully work that feels like play. This is more likely when the activity fits one's enduring motives, dispositions, and interests; as a result, the activity is stably energized and is more likely to continue. Happiness enhancement may also take knowing about, and endorsing, the activity's SWB-enhancing purpose. If people do not want to become happier, do not believe it is possible, or are not willing to invest the energy, then their SWB is unlikely to change (notably, however, too strong a focus on the SWB-enhancing effects of one's activity can also backfire; see Schooler, Ariely, & Loewenstein, 2003). Happiness enhancement also takes attending to the timing of activity; one should not over-regularize it or do it in small chunks, but perhaps instead should try to do it all out from time to time, in order to derive the maximal boost from that activity. Finally, happiness enhancement also requires varying one's activity. In order to maintain the flow of novel positive experiences, one must continually adjust and alter factors such as when, how, where, and with whom one does the activity, thereby forestalling the effects of hedonic adaptation.
Perhaps the most important take-home message of the research so far, however, is that even if one does everything right, the new intentional activity may still not work. Content matters, and even the best of intentions will fail to produce enhanced SWB if the activity itself does not provide an opportunity for positive experiences and personal growth (Sheldon et al., 2004). Activities that have thus far received experimental support include contemplating best possible selves, cultivating gratitude, being kind, replaying happy life events, savoring daily experiences, and employing one's strengths, among others (Dickerhoof et al., 2007; Lyubomirsky, Sheldon, & Schkade, 2005; Lyubomirsky, Sousa, & Dickerhoof, 2006; Seligman et al., 2005; Seligman, Rashid, & Parks, 2006; Sheldon & Lyubomirsky, 2006b; Tkach, 2005). In addition to revealing that the type of activity matters, the self-selection X activity-type interaction found by Dickerhoof et al. (2007) also suggests that the ‘sustainable happiness interventions’ conducted thus far are not just finding placebo and demand effects; even when people are highly motivated, if their assigned activity is uninspiring, there will be no gain in SWB. Thus, both the ‘why’ and the ‘what’ of activity appear to matter for SWB (see Sheldon et al. 2004, for a similar conclusion concerning personal goal pursuit).
To return to the questions posed in the title of this article: Yes, it may be possible to become a happier person, and emerging research is supplying concrete and experimentally validated information on how this can be accomplished. It takes effort and attention, but it can be done.