The present study, conducted in a relatively large sample of adults, revealed positive associations amongst the MiL dimensions as well as with CR measures. However, not all of the studied variables explained the self-reported health outcomes to a similar degree. Negative affectivity, including depression, anxiety, and stress, only showed an association with the MiL dimensions. By contrast, self-perceived cognitive function also presented an association with CR that was partially mediated by PiL and SoC. In the following sections we discuss these findings in the light of previous literature and highlight its relevance for further studies in the CR field.
Relationship between the MiL dimensions
Our correlation analyses revealed significant associations amongst the MiL components. Hence, although our confirmatory factor analyses indicated that SoC, PiL, and EwL represented partly independent psychological dimensions, they are strongly interrelated. These positive associations are not surprising given the findings of previous studies. For example, Waytz et al. [
81] reported a strong relationship between the presence of meaning, PiL, and satisfaction with life in young individuals. Regarding the engagement and fulfillment dimension, in the original study validating the EwL scale, the authors reported positive associations between this dimension and the six components of the PWB scales including PiL [
76]. Furthermore, conceptually, one of the dimensions of SoC, meaningfulness, refers to its motivational component, as it explores the extent to which life makes sense emotionally and whether the demands encountered are worth the commitment and energy so that they are seen as challenges instead of burdens. Hence, this dimension of SoC has elements of both the motivational (purpose) and the affective (fulfillment) aspect of MiL (i.e., [
39]). Our observations appear to be in accordance with the conclusions of Martela and Steger [
49], who, after reviewing the theoretical and empirical differences and commonalities between the different MiL dimensions, identified separate but synergic and necessary connections in the development of a sense of meaning.
MiL, CR, and self-reported affective and cognitive outcomes
Regarding the affective status outcome, we observed a direct effect of all three MiL dimensions, with higher scores associated with a better self-perceived mental state. The EwL scale used is composed of two subscales, which reflect the recognition and knowledge of personal values (valued living) and the sense of fulfillment in life as a consequence of recognizing and living in accordance with such values (life fulfillment). In this regard, higher EwL scores correlating with lower perceived negative affectivity are consistent with clinical investigations demonstrating the efficacy of therapeutic approaches that aim to increase value-oriented behaviors in the treatment of anxiety [
75]. Similarly, the association of PiL and SoC with affective status is not surprising since general conceptualizations of these two components specifically include the notion that they ‘buffer’ against stressors. Our findings are also in accordance with empirical studies showing a positive direct or moderating effect of these dimensions on mental health outcomes including stress, anxiety, and depression or depressive symptoms (e.g., [
24,
84,
85]).
We observed that PiL and SoC positively correlated with self-reported cognitive function. CR was also associated with perceived cognitive function, but this was partially mediated by SoC and PiL. The association between PiL, CR, and cognitive function in our subjects has some parallels in the literature. For example, Lewis et al. [
42] reported that PiL scores were positively associated with executive functions, memory, and general cognitive performance across the adult lifespan (32-84 years; see also [
84]). Furthermore, McKnight and Kashdan [
51] concluded that a certain degree of abstract capacity, insight, and planning explains interindividual differences in purpose formation, such high-order cognitive processes having also been previously associated with CR estimates [
57]. Moreover, in addition to the aforementioned studies correlating high PiL with lower incidences of AD and a reduced risk for MCI [
17], purpose seems to act through a similar ‘compensatory mechanism’ as CR. For example, PiL has been reported to moderate the relationship between AD pathology and cognition, as participants with higher levels of PiL were observed to show relatively better levels of cognitive performance at higher levels of neuropathology than those with lower levels of PiL [
18]. Other studies have indicated that psychological well-being (where PiL is one of the components) may serve as a buffer against adverse health events, specifically interacting with the educational background of the participants, which is a commonly used CR proxy variable. For example, in the Midlife in the United States (MIDUS) survey, Ryff et al. [
67] observed that a stable high level of well-being predicted better perceived health, lower incidences of chronic conditions and symptoms, and more preserved instrumental daily activities over time (see also [
41,
53,
55]). Notably, the positive effects of maintaining a high level of well-being on health outcomes were especially pronounced in those with a lower level of education.
In accordance with the conceptualization that PiL buffers against stressors, thereby providing an alternative compensatory mechanism to protect cognition, Wilson and Bennett [
83] concluded, after reviewing the findings of the Rush Memory and Aging Project, that ‘for the most part, psychosocial measures have not been correlated with neuropathologic changes traditionally associated with dementia in old age’. However, greater levels of PiL have been reported to be associated with fewer subcortical gray matter lacunar infarcts in older adults. This association persisted after adjusting for vascular risk factors, other psychosocial risk factors (i.e., depression, adverse childhood experiences, and loneliness), and physical activity, which have all been previously associated with PiL [
86,
87]. Hence, these findings indicate a possible neuroprotective effect of PiL on brain health rather than a compensatory mechanism, a topic of discussion that is currently highly relevant in the CR field (see [
9]). It should be noted, however, that the study by Yu et al. [
86,
87] did not confirm that a direct protective effect conferred a cognitive advantage in individuals with high PiL as there was no association between the measure (i.e., lacunar infarcts) and a cognitive outcome. Overall, given the previous and current observations of associations between CR and PiL, future studies using biological correlates (e.g., multimodal neuroimaging) should investigate if and how some MiL dimensions contribute to brain resistance beyond the previously reported resilience mechanisms (see a perspective of these terms in [
10]).
Associations between SoC and cognitive outcomes and how this may relate to concepts classically linked to CR (‘compensation’) have not been thoroughly investigated. However, a Finnish study [
62] reported that SoC was positively associated with both measurable and self-reported cognitive function, with SoC mediating a positive impact of cognitive status on social and mental health (i.e., depression and anxiety). In very old adults, Lövheim et al. [
46] observed that decreased cognitive function paralleled a declining SoC over time, while Boeckxstaens et al. [
15] reported that a high level of SoC correlated with baseline intact cognition. Moreover, SoC has been found to be associated with education, a common proxy for CR [
27], while leisure-time physical activity, another CR proxy, has been reported to have a positive effect on enhancing SoC [
54]. In addition, individuals with a high SoC have been shown to exhibit better cognitive performance when encountering an experimental stressor [
37]. Altogether, these results agree with our observed associations between SoC and self-reported cognitive status and the involvement of CR in this relationship. None of the aforementioned studies have investigated the biological, in particular cerebral, mechanisms through which high levels of SoC protect cognition in older adults and whether CR moderates this effect. Further longitudinal data from the BBHI study should help to clarify these issues.
Our investigation had several limitations. First, although we included a large sample of participants, the cross-sectional nature of our study meant that the direction of causality of the observed associations could not be determined. Recent longitudinal studies have observed that the stability of the different dimensions of well-being, rather than baseline measurements, correlates with better self-reported [
67] and objectively measured [
36] health outcomes. Furthermore, our regression analyses might have been limited by the fact that we applied linear regression models to non-normally distributed data. Moreover, we did not include an objective measure (i.e., neuropsychological performance) of cognitive performance and some psychological/psychosocial aspects, such as personality traits or social engagement previously linked to brain health status, were not included in our analyses. However, all of these issues should be addressed when the BBHI study undertakes assessments of the participants, which will include biological characterizations (i.e., neuroimaging and biomarkers). Furthermore, specific questionnaires measuring MiL dimensions other than the ones explored here are available in the literature (e.g., [
40,
72]), such as the Meaning in Life Questionnaire [
72], which measures purpose and coherence, but not the EwL dimension. Finally, we did not include measures to assess the germane concepts of meaning, particularly that of spirituality, which have been considered in several reports [
59].