A strong determinant of mood happiness is the influence of internal mechanisms of homeostatic control to retain jurisdiction of SWB when confronted by a threat (Cummins, 2017).  These internal mechanisms not only assist in retaining control of SWB, but they also restore the dominance of HPMood after an affective experience moves an individual’s SWB away from their set-point (Cummins & Wooden, 2014).  These internal homeostatic control mechanisms include habituation and adaptation, and three cognitive buffers conceptualized as self-esteem, perceived control and optimism. 

Habituation and Adaptation

Nomenclature around habituation and adaptation is easily confused.  Both terms refer to an organism becoming desensitized to new stimuli, although they have differential influences concerning the operation of SWB homeostasis.

Habituation is defined by Thompson (2009) as a ubiquitous neurophysiological phenomenon whereby exposure to a repeated frequent stimulus decreases subsequent responsiveness within organisms.  For example, a new proximal sound may draw an individual’s attention; however, the individual becomes accustomed to this audible stimulus over time.  Subsequently, attention and response to the noise diminish.  This diminishing response is habituation. 

Habituation is adaptive, involving primitive learning mechanisms evolved to conserve resources in the presence of predictable stimuli (Humphrey, 1933).  Therefore, as an internal reflexive mechanism of homeostatic control, habituation mitigates minor predictable positive or negative challenges to SWB (Cummins, 2018).  Consequently, under normal conditions, an individual’s optimal functional state remains mostly unaltered as minor agitations to SWB are corrected by habituation.  SWB is thus effectively maintained around its set-point.

Adaptation, like habituation, involves an organism becoming desensitized to stimuli.  However, the nature of the stimulus differs between these constructs.  While habituation is a response to a repeated frequent stimulus, adaptation refers to the process whereby an organism adapts to a continuous, unchanging stimulus (Helson, 1964; Luo, 2019).  Within SWB homeostasis, adaptation is a fundamental psychological buffer. 

Over time, adaptation allows people to tolerate a reduced negative or positive reaction to changed circumstances (Andrews & Withey, 1976).  For example, when Brickman et al. (1978) compared happiness levels between people who won the lottery and people who became paraplegic following trauma 18 months after the event, despite an initially large discrepancy, happiness levels for each group had returned to their baseline levels. One explanation for these findings is the adaptation process described by Helson’s (1964) adaptation level theory. 

Adaptation level theory posits that when an individual’s assessment of their current level of stimulation is higher (as was the case for Brickman’s (1978) lottery winners) or lower (Brickman’s (1978) paraplegic participants) than usual, an unconscious process is enacted to dampen the valence of the emotional response.  This is referred to by Helson (1964) as a shift in adaptation level whereby the novel situation is now re-categorized as normal.  At this point, the individual has become accustomed to the permanent stimulus.

Despite little attention given within the corpus of SWB literature to confirming Helson’s (1964) theory, and the insufficiency of non-significant correlational findings to identify the causal mechanisms of adaptation, SWB homeostasis has annexed adaptation level theory.  This is due to proponents of SWB homeostasis theory finding within adaptation level theory an explanatory model for the observed stability of SWB levels in the face of permanent positive and/or negative changes in an individual’s life circumstance.  They state, “It is clear that adaptation to altered circumstances of living does occur through one means or another, and that such processes are involved in maintaining the set-point-range of SWB” (Cummins & Lau, 2004, p. 288).

 In addition to habituation and adaptation, internal homeostatic control mechanisms include the cognitive buffers: self-esteem, optimism and perceived control (Cummins, 2018; Taylor et al., 2012).  This section will define each, then report evidence for their association with SWB and describe the most common measures for gathering self-esteem, optimism and perceived control data.  The conclusion of the section describes their role as a buffer within SWB homeostasis theory.

Self-esteem

Literature is replete with studies concentrated on self-esteem.  For example, in 2019, a search of the PsychInfo database yielded 50,781 results for self-esteem compared with 7,891 for SWB.  Within many of these studies, self-esteem is described in line with Rosenberg’s (1979) definition as an attitude of self-worth held by an individual towards an object.  In this case, the object is one’s self.  According to Rosenberg, an individual’s attitude towards themselves gives rise to either positive or negative feelings. Maintaining positive feelings towards oneself is fundamental and ubiquitous to human beings (Kaplan, 1975; Rosenberg, 1979). 

Self-esteem has been theoretically connected to SWB.  For example, Diener et al. (2015) locate SWB within his theory of positive mood offset (a term analogous to the mood component of Cummins’ HPMood and Russell’s core affect ) by conflating positive mood offsetwith Leary’s sociometer theory of self-esteem (Leary, Tambor, Terdal, & Downs, 1995).  This conflation explains why, universally, people feel and maintain mildly positive moods in the absence of strong emotional events by connecting an individual’s mood with their personal relationships and self-esteem. 

For example, according to Diener et al. (2015), when close relationships elicit feelings of being valued by another, an individual’s self-esteem and positive mood are sustained.  By contrast, exclusion from relationships can lower someone’s sense of worth and, if left unaddressed, will eventually negatively impact mood.  In addition, Diener et al. (2015) also include adaptive responses to lowering self-esteem levels within their explanatory model.  Behavioural adaptations are activated that increase the likelihood of future connections with other people; thus, when these behaviours produce the desired result, restored personal connection to others, self-esteem is increased, which moderates the return of an individual’s normal level of positive mood.

In addition to theory, empirical associations between self-esteem and SWB are well documented (Hajek & König, 2019).  While Diener et al. (2015) may assert that “self-esteem can be considered a form of subjective well-being” (p. 236), these empirical findings suggest while self-esteem and SWB share similar properties, empirical findings support both their association and their essential differences.  For example, Campbell (1981) showed self-satisfaction was the strongest predictor of SWB (.55) compared to other variables such as salary (.48), family (.45) and friends (.39).  In addition, Argyle and Lu (1990) reported that self-satisfaction more strongly predicts life satisfaction than personality variables such as extraversion.  Similarly, Diener and Diener (1995) identified self-esteem as the strongest predictor of SWB in their between-country study, reporting significant correlations in 29 of 31 countries, the highest being America at .60.  Assessing the effect size of these findings from the report is opaque.  However, these authors, recognizing the need for this information, state, “Our results point to the need for examining whether effect sizes systematically vary across cultures” (Diener & Diener, 1995, p. 661). 

In reviewing the cross-cultural findings from Diener and Diener (1995), Cummins and Nistico (2002) noted that the strength of associations between SWB and self-esteem were stronger in individualistic compared to collectivist cultures (see Na et al., 2010 for a cautionary note regarding generalising cultural group distinctive to individuals within the group).  According to Diener and Diener (1995), this is because, within individualistic contexts, such as Australia, people are socialized to attend to their own inner attributes. Varnum, Grossmann, Kitayama, and Nisbett (2010) agree, noting cultural contexts validating independence “tend to emphasize self-direction, autonomy, and self-expression” (2010, p. 9).  Thus, it is likely that people within individualist cultures primarily utilise their unique attitudes, emotions, and cognitions when making SWB judgments, resulting in stronger correlations between self-esteem and SWB (Lai, 2015).  By contrast, collectivists are socialized to understand themselves as fulfilling duties within the community (Tam, Lau, & Jiang, 2012) and attending to harmony with other community members (Grossmann & Na, 2014; Varnum et al., 2010).  Thus, feelings about one's self are less salient to the satisfaction-with-life construct than the organized social order, evidenced by lower correlations between self-esteem and SWB. 

Finally, Diener and Diener (1995) also suggest that the differential importance of self-esteem in collectivist and individualistic nations is due to the socialization of affect.  This is premised on their findings that students in the United States believe positive affect is normative, whereas students in Korea and China included negative affect.  Therefore, these authors hypothesise that “life satisfaction may be based more on positive feelings in individualistic nations, [than] in collectivist nations [where] life satisfaction might be influenced by a more prevalent negative focus and therefore be more dependent on how many problems and social conflicts the person faces” (p. 662).

Whatever the cause, Diener’s (1995) observation of different associations between self-esteem and SWB as a function of whether people are participating in individualistic or collectivist cultures is also supported by (Lai, 2015) and has major implications.  At the very least, these observations demand a cautious approach to gathering data from collectivist cultures.  Researchers ought to include methodological safeguards limiting the introduction of bias from strong cultural forces and deploy analytic strategies such as multi-group confirmatory factor analysis to evaluate the effectiveness of these strategies.  These methods would constitute a minimum before any valid inferences could be drawn from gathered cross-cultural data using Western scales.

The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) is the most widely employed instrument for measuring self-esteem.  It is favoured for its brevity, manifold translations and reported invariance across cultures.  The questionnaire asks people, regarding their current feelings, to indicate their level of agreement (measured on a 4-point scale “strongly agree, agree, disagree, strongly disagree), with the following 10 statements:

1.     On the whole, I am satisfied with myself.

2.     At times I think I am no good at all.

3.     I feel that I have a number of good qualities.

4.     I am able to do things as well as most other people.

5.     I feel I do not have much to be proud of.

6.     I certainly feel useless at times.

7.     I feel I am a person of worth, at least on an equal plane with others.

8.     I wish I could have more respect for myself.

9.     All in all, I am inclined to feel that I am a failure.

10.  I take a positive attitude towards myself.

A self-esteem score is obtained by summing the five positively worded items and five negatively worded items (reverse scored).  High scores indicate robust levels of self-esteem.  The measure has sound psychometric properties with test-retest correlations ranging between .82 to .88, and a Cronbach's alpha reported at .86 (Mellor et al., 2008). 

However, Song, Cai, Brown, and Grimm (2011) and Chao, Vidacovich, and Green (2016) report differential item functioning for the negatively worded items between Western and cross-cultural contexts.  These authors argue that the differential item functioning is suggestive of the increased cognition required in responding to negatively worded questions on an agree-disagree scale and impacts the construct validity by consistently producing a bi-factorial structure (Quilty, Oakman, & Risko, 2006; Salerno, Ingoglia, & Lo Coco, 2017). As a result, they recommend researchers only employ the five positively worded items in their scales “as the advantages of including both positively and negatively worded items seem to be offset by these methodological issues” (p. 114). 

Despite the problems identified with retaining positive and negatively worded items, except for single-item measures of global self-esteem (Brailovskaia & Margraf, 2018), a search in 2019 of Medline Complete, MeSH, PsycINFO, socINDEX and Google Scholar failed to locate studies employing only a positively worded version of the RSES.  By contrast, to reduce the possible introduction of bias from the negatively worded items, the Australian Unity Wellbeing Index survey regularly employs the five positively worded items RSES as a measure of self-esteem. 

 Optimism

Optimism is defined by Scheier and Carver (1985) as the expectancy held by individuals that future life outcomes are likely to be favourable.  Further, these authors argue that an optimistic attitude regulates behaviours such as an initial willingness to engage with fearful activity and continued engagement.  By contrast, those with lower levels of optimism are less likely to engage and quicker to disengage with fear-inducing activity.  As a result, it is hypothesised that higher levels of optimism translate to successfully engaging in more positively valenced outcomes, increasing an individual’s overall SWB (Diener et al., 1999).

A positive association between optimism and SWB (r = .61) is reported by Dember and Brooks (1989), who measured optimism within 106 students by items such as “I generally look at the brighter side of life.”  In addition, similar empirical links between optimism and SWB have been reported by Lucas, Diener, and Suh (1996), reporting a correlation of .60 and Compton (2000), at .52.

To measure dispositional optimism and pessimism, a re-evaluation of the original 12-item Life Orientation Test (LOT, Scheier & Carver, 1985) by Scheier, Carver, and Bridges (1994) resulted in the production of the shortened version, referred to as the Revised Life Orientation Test (LOT-R), correlating .95 with the original LOT.  The preeminence of the LOT-R, as a measure of optimism within SWB literature, is well attested, returning 14,900 citations through Google Scholar in 2019.  The scale contains ten items consisting of three positively worded items and three negatively worded items, plus four filler items that are not scored. 

Despite the prominence of this 10-item scale, a reasonable argument is made within the literature to justify shortening the scale further.  For example, an analysis of the LOT-R by Maher and Cummins (2001) suggested that filler items, designed to highlight response sets, add unnecessary length to questionnaires and response sets are easily identified via other methods during data cleaning, which researchers remove before analyses.  Further, Scheier and Carver (1985) designed the LOT-R to measure optimism; therefore, the negatively worded statements assessing pessimism are likely redundant. 

Empirical justification for the redundancy of items assessing pessimism is supported by the CFA analyses of the LOT-R by Herzberg, Glaesmer, and Hoyer (2006).  This study, examining data collected from 46,133 people, confirms that the two-factor solution for the LOT-R (CFI = .99; RMSEA = .04) is a superior fit to these data when compared to a single factor model (CFI = .62; RMSEA = .21).  Further, studies confirming the reliability of the optimism dimension, including Reilley, Geers, Lindsay, Deronde, and Dember (2005), reporting Cronbach’s alpha’s of .83 and .82  in studies consisting of 204 and 273 respondents respectively, and, Chang, Maydeu-Olivares, and D'Zurilla (1997) who also report a Cronbach’s alpha of .91 for the optimism dimension of the LOT-R.  Finally, correlations supporting the discriminant and convergent validity of the LOT-R’s optimism dimension are also reported within these studies by Reilley et al. (2005) and Chang et al. (1997).

As a result, researchers such as Vautier, Raufaste, and Cariou (2003) are justified when they state, “there is no empirical necessity for hypothesizing that the dispositional optimism construct must be split into optimism plus pessimism” (p. 390).  Therefore, following suggestions by Maher and Cummins (2001) and  Vautier et al. (2003), only the three positively worded statements from the LOT-R were included throughout the current thesis.  These statements are:

1.     In uncertain times, I usually expect the best.

2.     I'm always optimistic about my future.

3.     Overall, I expect more good things to happen to me than bad.

Perceived Control

Perceived control is defined within the current thesis as the extent to which individuals believe they are in control of their lives.  By contrast, based on research by Peterson (1999), perceived control has also been defined as how people behave to augment good outcomes for their lives (Blore, 2008).  The distinction is pertinent. 

Perceived control, operationalised according to how people behave to maintain control of their lives, has informed the selection of scale items measuring control on most Australian Unity Wellbeing Index surveys conducted before and after the 23rd edition.  For these surveys, various control strategies are derived from Heeps (2000) and based upon theory originating from Rothbaum, Weisz, and Snyder (1982). 

The Australian Unity Wellbeing Index questions pertaining to control strategies begin with the statement "When something bad happens to me" and are followed by statements such as "I ask others for help or advice" and "I remind myself something good may come of it." Respondents represent their level of agreement on an 11-point scale ranging from 0 (“strongly disagree”) to 10 (“strongly agree”).  These items require participants to consider the strategies they use to regain control when ‘something bad happens,’ with the first response tapping a behavioural tendency that enlists external resources to maintain control, and the second, describing a preferred internal cognitive response utilised to restore control.  The statement "When something bad happens to me" and the two options offered to respondents answering the Australian Unity Wellbeing Index tap their preferred way of coping, not the degree to which they feel they are coping.

When the definition of perceived control is operationalised to measure the strategy employed to regain control, the extent to which an individual feels they are currently in control of their life is not tapped.  In relation to SWB, Diener et al. (2015) note the primacy of these feelings as the catalyst for subsequent behaviours that an individual engages to maintain control and their subsequent set-point for SWB.   Similarly, within SWB homeostasis theory, there is a theoretical relationship between feelings of control, subsequent behaviours engaged in maintaining these (as reflected by the questions posed within the Australian Unity Wellbeing Index listed within the prior paragraph) and the buffering of threats to SWB. 

There exist discrete scales to measure each.   For example, the one used by the Australian Unity Wellbeing Index to assess preferred behavioural responses to the loss of control, and another, a revised version of Pearlin and Schooler’s scale (1978), to measure perceived control, as defined within the current thesis.  Pearlin and Schooler’s (1978) seven original items are:

1.     I have little control over the things that happen to me.

2.     There is really no way I can solve some of the problems I have.

3.     There is little I can do to change many of the important things in my life.

4.     I often feel helpless in dealing with the problems of life.

5.     Sometimes I feel that I'm being pushed around in life.

6.     What happens to me in the future mostly depends on me.

7.     I can do just about anything I really set my mind to do.

A revised version of Perlin and Schooler’s original scale (1978), employed within the 23rd Australian Unity Wellbeing Index, reduced the number of scale items from seven to five by removing the positively-worded items.  The rationale for reducing items includes producing a scale with consistency regarding its valence, and, avoiding likely bias introduced by differential item functioning from including both positive and negatively worded items.  Items retained by the 23rd Australian Unity Wellbeing Index are:

1.     I have little control over the things that happen to me.

2.     There is really no way I can solve some of the problems I have.

3.     There is little I can do to change many of the important things in my life.

4.     I often feel helpless in dealing with the problems of life.

5.     Sometimes I feel that I'm being pushed around in life.