Quality of life’ is a lovely phrase. It embraces feel-good words like happiness, flourishing and human potential. QOL also sounds aspirational, so everyone should have it, and more is better. Because of these connotations, the term has high marketing appeal. The promise of enhanced QOL is used to sell diamonds, real estate, cat food and politics. It has spawned countless self-help books, courses and gurus with a mission to enhance your wellbeing. Most famously, within our own discipline, it is linked to the rise of ‘positive psychology’.
Our discipline is enmeshed with fostering life quality. A core aim is to assist people in their wise use of personal resources, that they may thrive and overcome adversity, to enhance the quality of their life. Understanding the science that informs this aim falls largely within the purview of Subjective Wellbeing (SWB). This is the subjective dimension of the QOL construct, contrasting with the objective dimension, represented by such traditional indices as money and medical health. SWB is the self-perception of our own wellbeing.
Research into SWB tackles fundamental questions concerning how humans manage to keep themselves feeling normally positive and motivated. Knowledge here informs psychological practice because of clear links to psychopathology when the normal wellbeing systems fail. So, SWB has become a popular area for study. It not only holds promise for understanding the human condition, but also involves many of our discipline’s favourite constructs: personality, perceived control, self-esteem and optimism, as well as being an interesting amalgam of cognition and affect. So, why should we be cautious about accepting how such an engaging construct can be used to foster life quality? Let me list the ways.
Caution 1: Flawed literature
The central core of our discipline is to understand constructs through the application of scientific method, and then to apply such understanding to evidence-based, psychological practice. This generally works well in areas such as learning theory and the Poggendorff illusion. It works less well when the topic is a feel-good construct. The lurking peril is that the beguiling nature of topics such as ‘human flourishing’ blunt our scientific scepticism, so carefully drilled during university training, causing un-critical acceptance of weak evidence. So, the first problem is the research base. Many of the claims made in the literature, especially regarding the efficacy of techniques to increase SWB and ‘happiness’, are seriously flawed. This is true even within many articles published in refereed journals. Be careful what you believe.
Caution 2: Terminological anarchy
It is a discomforting fact that the term QOL has no agreed meaning. For a start, there are disciplinary biases. For classical economists QOL is defined by wealth, which is considered proxy for happiness. Within medicine QOL is most commonly defined as the absence of medical symptoms, and when these are measured through patient self-reports, it is called Health Related QOL, which authors almost inevitably abbreviate to QOL. While within the social sciences, and most particularly within psychology, QOL is generally considered as Subjective Wellbeing.
Since systematic research into SWB has been underway for over 40 years, it might reasonably be expected that contemporary researchers, at least, are working in the context of an agreed definition. Curiously, and disappointingly, this is not so. Even the most senior academic researchers have differing interpretations of what is meant by the term. Not only is there no agreement as to the definition of SWB, this uncertainty also applies to the terms used to describe central SWB components, such as ‘happiness’, which is very disconcerting.
The term ‘happiness’ has been used in the contemporary literature as a blanket term to include all positive feelings about the self, as a synonym for SWB, as referring to average levels of positive and negative affect, and as a single affect within the classification system described by Russell’s circumplex model of affect. ‘Happiness’ is also used to imply different temporal durations, as a long-duration positive mood trait, or as a short-duration positive emotion. The latter is consistent with the term ‘happy’ in common English usage, which refers to a transient, positive state of mind that has been caused by a specific experience, such as a pleasant social interaction. In sum, the use of nomenclature within the SWB literature is shambolic, terminological anarchy rules and, in my view, this chaotic situation is the single most important impediment to theoretical advancement and the valid use of SWB in applied psychology.
Caution 3: Empirical bounty in a theoretical desert
The era when empirical, exploratory, atheoretical studies could advance understanding of SWB in non-trivial ways is over. Yet many such studies continue to be published, releasing in the process a cascade of results with no clear purpose or interpretation. Coupled with the nomenclature problems discussed above, this represents a great waste of effort and feeds confusion rather than comprehension. Understanding is only systematically advanced by the use of empirical studies to test theory, and there are many interesting theories to be tested. These involve such topics as cultural response bias, the composition of SWB, the functional advantages of high and low SWB, and the nature of the link between SWB and depression. There are also several over-arching theories which attempt to account for the character and behaviour of SWB, one of which is the theory of SWB homeostasis.
Subjective wellbeing homeostasis
The theory of SWB homeostasis (Cummins, 2010) proposes that, in a manner analogous to the maintenance of body temperature, the level of SWB is controlled by automatic neurological processes. Their purpose is to maintain a sense of personal wellbeing that is mildly positive and activated. This feeling reflects the deep, stable, mildly-positive mood state that is the core of SWB, and which homeostasis seeks to defend. As a consequence of homeostatic maintenance, the level of SWB has some interesting characteristics.
SWB is normally stable and positive
The stability of SWB at the level of population sample mean scores is remarkable. Actually, it is extraordinary! Since 2001, a total of 28 surveys, each involving a new sample of 2,000 random, adult Australians, have been conducted as the Australian Unity Wellbeing Index. SWB is measured using the Personal Wellbeing Index and results are projected onto a standard 0 – 100 point scale. The total variation between these 28 survey mean scores is from 73.7 to 76.7, or just three percentage points.
The distribution of SWB is under genetic control
This positive stability in SWB is due to several factors, including the economic and political stability of this country. But the dominant cause of stability is proposed to be genetic. Each person has a set-point for their SWB that constitutes a genetically determined, individual difference. These set-points are normally distributed within the population over the range 70 to 90 points. Further, each set-point has a range of 18 – 20 points for each person, and it is proposed that homeostasis seeks to maintain SWB within this set-point range.
Low SWB is indicative of pathology
While SWB normally lies within its set-point range for each person, a strong emotional challenge will defeat homeostatic control. Values above or below this range indicate the dominance of a short-term emotional response. Persistently low values are associated with a high probability of depression.
So, how does homeostasis work to defend SWB against the unusually good and bad experiences of life? The answer we propose is that there are three levels of defence and we call these defensive systems ‘buffers’.
The first line of defence is behaviour. People are generally adept at avoiding unpleasant emotional challenges through established life routines. These help to make daily experiences predictable and manageable. However, strong and unexpected events will cause SWB to move out of its set-point range from time to time, as attention shifts to emotion generated by the event. Such deviations are usually brief. Adaptation to strong positive challenge is certainly predictable and fast. Disappointing as this may be, it is important to restore normal functioning. While elevated SWB is associated with nice features such as pleasure, generosity and enhanced friendliness, it also has the downside of inducing poor information processing, an exaggerated sense of control, and therefore an enhanced probability of risk taking. Successful adaptation to strong negative challenge is less certain. Here, adaptation depends on the balance between the strength and persistence of the challenge matched by the strength of personal resilience. Such resilience depends on the strength of external and internal resources.
External resources comprise the second line of defence, with the two most powerful an intimate relationship and money. While both are intuitive in this capacity, the role of money in fostering life quality is often misunderstood. It cannot, for example, be used to shift the set-point to create a perpetually happier person. Set-points for SWB are genetically determined so no matter how rich someone is, their average level of SWB cannot be sustained higher than the upper portion of their set-point range.
The true power of money to foster QOL is its potential to be used as a flexible resource. Money allows people to defend themselves against the negative potential inherent within their environment. Wealthy people pay others to perform tasks they do not wish to do themselves. Poor people, who lack such resources, have a level of SWB that is far more at the mercy of their environment. In summary, the external buffers assist with the homeostatic management of SWB, but they cannot prevent occasions when SWB moves outside its set-point range. When this occurs, the internal buffers are activated.
The internal buffers comprise protective unconscious and conscious devices designed to minimise the impact of personal failure on positive feelings about the self. The unconscious devices are immediately activated when SWB is threatened. For example, a negative feeling of social exclusion initiates an immediate and automatic emotion regulation process (non-conscious coping), in which positive emotions become highly accessible. The purpose is to defend SWB by counteracting the negative emotion.
Such a response can be conceptualised as a psychological ‘Band-Aid’ to help mend a small emotional wound. Larger injuries, such as may be inflicted by strong, prolonged negative challenges, require psychological bandages. These come in the form of conscious coping devices which have been variously described by such terms as secondary control and positive reappraisal. These coping devices protect SWB by altering the way we see ourselves in relation to homeostatic challenge, such that the negative potential in the challenge is deflected away from the core view of self. The ways of thinking that can achieve this are highly varied. For example, one can find meaning in the event (‘God is testing me’), deny responsibility for the failure (‘it was not my fault’) or regard the failure (dropping a fragile object) as unimportant (‘I never did like that old vase anyway’).
In summary, the combined behavioural, external and internal buffers ensure that our genetically provided, normal level wellbeing is robustly defended. The integrity of this defence is essential to fostering life quality by allowing acute variations in emotional experience to be coupled with chronic stability of SWB. But what is it, actually, that is being defended?
What is homeostasis defending?
Most contemporary theorists regard the composition of SWB, measured through questions of ‘satisfaction’, to involve both affective and cognitive components. However whether, as is commonly claimed, SWB is dominantly cognitive, is moot. To the contrary, recent research weighs the balance strongly in favour of affect as the central element, in the form of a deep and stable positive mood state. We refer to this as Homeostatically Protected Mood (HP Mood: Cummins, 2010) and propose that it has the following characteristics.
- HPMood is a genetically-based, neuro-physiologically generated, stable, positive-activated mood. Being a trait, the presence of HPMood is not tied to any experiential causative agent in the manner of an emotional response. Instead, it represents the most basic way each individual can sense themselves in an abstract and personal way. HPMood may be described as analogous to felt body temperature in that it is always there, can be measured under the right circumstances, and exists without words to describe it. In conformity with the circumplex model of affect, HPMood comprises a blend of hedonic (pleasant) and arousal values (activation).
- The traditional view of SWB is that it is strongly influenced by personality. This now appears to be incorrect. As demonstrated within three separate reports from our research group, structural modelling reveals that HPMood dominates both personality and SWB. In other words, personality correlates with SWB only because both variables are being influenced by HPMood.
- SWB is highly saturated with HPMood. Hence, under benign life circumstances, HPMood is a reasonable proxy measure of SWB. This is most evident when SWB is measured by ‘How satisfied are you with your life as a whole?’. The abstract and personal nature of this question means that people formulate their answer using the level of HPMood as information.
- While SWB is a proxy for HPMood under benign conditions, in most real life circumstances the degree of correspondence is certainly imperfect. HPMood is proposed as a genetically-produced mood that never changes. SWB on the other hand, measured by scales such as the Personal Wellbeing Index, is contaminated by affect from other sources, such as momentary emotional experience. Nevertheless, HPMood accounts for >60% of the variance in SWB in general Australian population samples.
- HPMood appears to comprise three main affects. These are dominated by a sense of contentment, flavoured with a touch of happiness and arousal. We propose that these affects combine to provide each person with a unique level of felt positivity. This represents their ‘set-point’ and constitutes an individual difference between people. Each set-point is the level of HPMood that homeostasis seeks to defend.
In summary, HPMood is proposed as the basic psychological variable that is homeostatically defended. Due to its heavy saturation with HPMood, SWB is a reasonable proxy measure provided that homeostatic control is maintained. However, when homeostasis fails due to strong positive or negative challenge, SWB changes, HPMood remains steady, and so the degree of correspondence between them becomes unreliable. Understanding these relationships is central to understanding how best to foster life quality.
Using SWB to foster QOL
Subjective wellbeing is commonly used as a measure of intervention effectiveness and homeostasis theory makes determined predictions concerning its usefulness for this purpose. If these predictions are correct, they offer crucial signposts to the use of SWB both in academic research and in applied settings. If these predictions are not correct, they would demand a rethink of the homeostasis story, as told. This final section of the article addresses some of these predictions, first for theoretical understanding and secondly as a measure of outcome.
Fostering QOL through academic research
Academic research is about fostering understanding, so building theories that can be tested is the cornerstone of this endeavour. Three testable implications of homeostasis theory are as follows.
Implication 1: The crud factor
A large number of diverse scales have been used to measure SWB. So it might be reasonably expected that this would produce a corresponding diversity of results. Curiously, this is not so. Not only do different scales measuring SWB produce remarkably similar means, almost any of these scales obligingly provide a Cronbach alpha of .7 or higher. These scales also inter-correlate quite strongly, which allows authors to claim ‘validity’ for the metric they have generated. The data also predictably correlate with other self-report scales measuring such constructs as self-esteem, optimism, perceived control and personality at about .3 to .5. This is a dream scenario for researchers, but why? What is the origin and character of this predictable shared variance?
To my knowledge, the first person to show concern with this phenomenon was the redoubtable researcher Paul Everett Meehl. He condemned research in 'soft psychology' (e.g., survey research) for testing weakly substantive theories based on correlations and, at the age of 70 years, he vented his rage (Meehl, 1990). He recognised that such correlations were due to shared variance within self-report data, and he labelled this variance the ‘crud factor’. He wrote disparagingly: “In the social sciences—everything correlates to some extent with everything else” (p. 204). While Meehl’s statement somewhat exaggerates reality, confirmatory evidence can be observed in almost any correlation matrix involving variables derived from self-report survey data.
In an article using survey data (Lai & Cummins, 2012) we report the following correlations with general life satisfaction: self-esteem (.57), optimism (.42), primary control (.42) and secondary control (.36). After using HPMood as covariate these correlations are reduced to .22, .03, .13, and .01 respectively. We argue that HPMood is Meehl’s ‘crud-factor’. The shared variance is a consequence of the normal distribution of set-points within any representative general population sample. Because each set-point controls the level of HPMood, people with high set-points tend to score high on self-report variables, while people with low set-points tend to score low. Thus, the distribution of set-points becomes a distribution of systematic response bias within the range of 70 to 90 points. This distribution causes shared variance between self-report items, and so the scales comprising these items inter-correlate. The implication is that, in order to validly study such inter-relationships, the variance contributed by HPMood must first be removed.
Implication 2: Variable saturation with HPMood
Many scales measuring SWB comprise satisfaction responses to a list of personally-relevant items. Such lists are inevitably compiled without theoretical assistance. The implicit assumption is that items are equivalent in their capacity to represent SWB. Homeostasis theory challenges this assumption. The extent to which items are saturated with HPMood depends on two characteristics: distance from the self (proximal-distal) and degree of abstraction (abstract-specific). The items most highly saturated with HPMood are proximal-abstract (‘How satisfied are you with your life as a whole?’), those least saturated are distal-specific (‘How satisfied are you with the Prime Minister?’). Responses to the former are dominated by HPMood and therefore represent SWB. Responses to the latter are dominated by cognition and emotion attached to the item target. They weakly include HPMood and do not validly represent the SWB construct
Implication 3: Non-linearity
Homeostasis theory proposes that the relationship between SWB and any variable representing a level of challenge, is non-linear. This non-linearity is created by a plateau in the relationship as, in the face of increasing levels of challenge, homeostasis defends the line. The application of linear statistics to describe the relationship will produce a false impression of linearity, and the subtlety of the plateau will be obscured.
Fostering QOL by measuring SWB
It has been argued that, for people living in favourable environments, where SWB is generally within each set-point range, the strongest determinant of SWB is their set-point. Moreover, since the distribution of set-points within population samples is not influenced by advantageous resources, such as wealth or health, this is one reason that population levels of SWB in Australia are hard to change. However, this stability also depends on the resilience of each person, operationalised as the strength of their homeostatic system.
Resilience in the Australian population is generally high, and the means of homeostatic control has been described. As an important consequence, the influence of additional weak resources or challenges will normally have little impact on SWB. However, under conditions of chronic challenge by a strong emotional experience, homeostatic resilience is overwhelmed.
When homeostasis fails, people’s responses to SWB questions will be dominated by emotion generated by the challenging agent. Under such conditions, SWB will be highly sensitive to the power of resources and challenges since homeostasis is no longer operating as an effective buffering agent.
Implication 1: Baseline measurements
The most crucial understanding to be derived from this theory is that changes in SWB, due to challenges or resources, will be exquisitely sensitive to initial levels. If baseline SWB is within normal range, then it will appear more rigid than fluid. If SWB lies above range then, in the interval before the next measurement, adaptation alone will likely return it to baseline. If, however, SWB at baseline is below normal range, then SWB will show considerable sensitivity to the addition of positive resources.
Implication 2: Normative ranges
As a corollary of the above, interpreting baseline levels of SWB requires a knowledge of normative ranges. Such ranges, measured by the Personal Wellbeing Index, are available for the general Australian population and common demographic sub-groups (Cummins et al., 2012). Consulting such ranges is essential to determine whether groups or individuals, at baseline, are within the normal range.
Implication 3: Resilience
Perhaps we are using the wrong measures to identify the influence of resources and challenges on the SWB of general population samples. In economically mature countries SWB is hard to change, for reasons that have been proposed. But studying changed mean scores may be missing a more powerful effect. The importance of adding or subtracting resources to normally functioning populations may not be found in altered levels of SWB, but rather in the increased homeostatic resilience such resources confer.
How to measure such resilience, without applying stress tests, is a challenge for the future.
The claims made for SWB homeostasis are coherent and far reaching. But how worthy is this hypothetical construct as a device to foster understanding? Certainly it is based on the best empirical evidence we have at our disposal and, to my knowledge, no killer-flaw has been discovered in the logic. However, some important aspects are glossed over rather too quickly. The cornerstone is the assumption of set-points for SWB. While we have convergent statistical evidence for their existence, the evidence will need to involve biology to become more assured. In the meantime, our search for dis-confirmatory psychological evidence continues. As I share with my students: we do not need more results consistent with homeostasis theory. We need to discover contrary evidence, and then be the first to know.
The author can be contacted at email@example.com
- Cummins, R. A. (2010). Subjective wellbeing, homeostatically protected mood and depression: A synthesis. Journal of Happiness Studies, 11, 1-17. doi: 10.1007/s10902-009-9167-0
- Cummins, R. A., Woerner, J., Weinberg, M., Collard, J., Hartley-Clark, L., Perera, C., Horfiniak, K. C. (2012). Australian Unity Wellbeing Index: Report 28.0. The Wellbeing of Australians and the impact of marriage. Melbourne: Deakin University. ISBN 978-1-74156-174-6
- Lai, L. C. H., & Cummins, R. A. (2012). The contribution of job and partner satisfaction to the homeostatic defense of subjective wellbeing. Social Indicators Research doi: 10.1007/s11205-011-9991-6
- Meehl, P. E. (1990). Why summaries of research on psychological theories are often uninterpretable. Psychological Reports, 66, 195-244.
- Additional evidence for the various claims discussed in this article can be found in the following publications.
- The interaction between affect and cognition
Forgas, J. P. (2008). Affect and Cognition. Perspectives on Psychological Science, 3(2), 94-116.
- The circumplex model of affect
Yik, M., Russell, J. A., & Steiger, J. H. (2011). A 12-point circumplex structure of core affect. Emotion, 11(4), 705-731.
- Critique of positive psychology
Cummins, R. A. (2013). Positive Psychology and Subjective Wellbeing Homeostasis: A critical examination of congruence. In D. Moraitou & A. Efklides (Eds.), Quality of Life: A Positive Psychology Perspective (pp. 67–86). New York: Springer.
- Subjective wellbeing and health psychology
Cummins, R. A. (2012). The Relationship between Subjective Wellbeing and Health. In M. Caltabiano & L. Ricciardelli (Eds.), Handbook of Health and Well-Being (pp. 101-111). Oxford: Wiley.
- A taxonomy of QOL terminology
Diener, E. (2006). Guidelines for national indicators of Subjective well-being and ill-being. Journal of Happiness Studies, 7, 397-404.
- Subjective wellbeing cultural response bias
Lai, L.C. H., Cummins, R. A. & Lau, A.L.D. (in press). Cross-cultural differences in subjective wellbeing: Cultural response bias as an explanation. Social Indicators Research.
- Demonstration of set-points
Cummins, R. A., Li, N., Wooden, M., Stokes, M. (in press). A demonstration of set-points for Subjective Wellbeing. Journal of Happiness Studies.
- Using mood as information
Schwarz, N., & Clore, G. L. (1996). Affect and phenomenal experiences. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 433-465). New York: Guilford Press.