The Role of Time Perspective and Mindfulness on Life Satisfaction in the United States of America, Spain, Poland and Japan: A Cross-Cultural Study

Abstract

We assessed the cross-cultural role of Time Perspective (TP) tendencies [Past Positive (PP), Past Negative (PN), Present Hedonistic (PH), Present Fatalistic (PF), and Future (F)], the Deviation from a Balanced Time Perspective (DBTP) profile, the Deviation from a Negative Time Perspective (DNTP) profile, and mindfulness on life satisfaction (LS). The sample consisted of psychology undergraduate students (N= 867, M_AGE= 20.19, SD= 3.417) in four countries: USA, Spain, Poland and Japan. We used a 17-item short version of the Zimbardo Time Perspective Inventory (ZTPI), the Mindful Attention Awareness Scale (MAAS), and the Satisfaction with Life Scale (SWLS) in all countries. For ensuring measurement invariance, we conducted pairwise CFAs for the ZTPI-17, MAAS and SWLS. Regression analyses showed that PN predicted decreased LS in Poland and Japan. PP predicted increased LS in Spain. F predicted increased LS in Poland. DNTP predicted decreased LS in Poland. Mindfulness predicted decreased LS in Japan and increased LS in USA, Spain and Poland. Moreover, mediation analyses revealed that the DBTP partially mediated the relationship between mindfulness and LS in Spain and USA. The DNTP partially mediated the relationship between mindfulness and LS in Spain, Poland and Japan (opposite direction). The findings suggest that the association of TP, mindfulness and LS differs across the investigated countries as a function of culture.

Introduction

Psychological processes have a great influence on culture, and culture affects basic psychological processes of individuals. Cross-cultural psychological research examines similarities and differences between cultures, based on “socially shared practices, norms, values, and other mental events that are loosely organized around some common theme” (Shore, 1998; Triandis, 1989). The majority of well-being studies conducted in diverse locations throughout the world suggest that human conceptualizations of well-being and life satisfaction (i.e.: the cognitive aspect of subjective or hedonic well-being) vary across nations (Diener et al., 2010). For instance, some psychosocial factors that have consistently been shown to influence international variations in well-being include culture (Oishi, 2010), socio-economic status (Easterlin & Sawangfa, 2010; Layard et al., 2009; Tella & MacCulloch, 2008) and social context (Helliwell et al., 2009). In the case of life satisfaction (LS) several cross-cultural research studies have found that individuals in the United States of America (USA) rank lower in LS compared to various European countries with lower per capita incomes (Alesina et al., 2004; Blanchflower & Oswald, 2004; Di Tella et al., 2003). As these studies indicate, psychosocial factors indeed influence variability in reports of well-being and LS across nations. Therefore, we have decided to conduct cross-cultural research, in order to highlight similarities and differences between countries.

Time Perspective

A psychosocial factor that has recently been studied in relation to well-being and LS is Time Perspective (Olivera-Figueroa et al., 2016). Time Perspective (TP) is defined as the cognitive-behavioral tendency through which personal, emotional and social experiences are assigned to different time frames (Kruger et al., 2008; Suddendorf & Corballis, 1997; Zimbardo & Boyd, 1999). This construct is considered to act as a cognitive frame that mediates situational responses and behavioral patterns through five tendencies measured by the Zimbardo Time Perspective Inventory (ZTPI): Past Positive; Past Negative; Present Hedonistic; Present Fatalistic; and Future (Zimbardo & Boyd, 1999).

Of the two TP tendencies measuring the past on the ZTPI, Past Positive has been characterized by a nostalgic view of the past, which promotes the retrospection of positive memories from the past, in an attempt to secure or even recreate them (Zimbardo & Boyd, 1999). When taken to an extreme, it can lead to a lack of focus in the present and future. In contrast, Past Negative refers to a pessimistic approach towards past events, characterized by a constant reconstructing of past negative events or traumas, as well as by an inability to move on from them (Zimbardo & Boyd, 1999). It is associated with feelings of regret and pain, as well as fear of repeating past negative experiences. When taken to an extreme, it can lead to an enhanced exposure to undermining interactions, social conflict, less social support, and briefer relationships in the aftermath of stress (Holman & Zimbardo, 2009).

Of the two tendencies measuring the present on the ZTPI, Present Hedonistic has been characterized by a focus on living in the here and now, in a manner focused on abstracting pleasure from the present (Zimbardo & Boyd, 1999). It denotes an intensive way of living, an avoidance of routine and boredom, and a tendency to take risks. When taken to an extreme, it can lead to risk behaviors through decreased consideration of future consequences, as well as by ignorance of healthy behaviors (Boniwell & Zimbardo, 2003). In contrast, Present Fatalistic refers to a pessimistic, passive way of perceiving life, characterized by the belief that one does not have the power to influence his own life (Zimbardo & Boyd, 1999). It is associated with low self-esteem, low external control, stagnation, helplessness, pessimism, passive way of perceiving life, and interpreting present situations through past negative experiences. When taken to an extreme, it can increase vulnerability toward stress-related problems like allostatic load (Bourdon et al., 2020), anxiety (Papastamatelou et al., 2015), and post-traumatic stress disorder (Zimbardo et al., 2012).

In the case of future, only one ZTPI tendency addresses this temporal orientation, identified simply as Future (Zimbardo & Boyd, 1999). This tendency has been characterized as an awareness about forthcoming goals and benefits associated with them. It is associated with setting realistic goals, internal control, and a proactive way of life. When taken to an extreme, it can lead to lack of spontaneity and present joy, excessive meticulousness and future anxiety (Zaleski, 1996).

As such, TP may be a manifestation of a dispositional style or an individual-difference variable. In this regard, TP is considered a psychological trait, through which personal and social experiences are assigned to different time frames, giving order, coherence, and meaning to life events (Zimbardo & Boyd, 1999). Importantly, TP has been shown to transcend cultural barriers, since a study conducted across 24 countries supported the existence of the described five TPs across all participating countries (Sircova et al., 2014).

Time Perspective’s Relationship to Life Satisfaction

In terms of TPs’ relationship to life satisfaction (LS), Past Positive has been shown to positively correlate with subjective happiness and LS, while Past Negative negatively correlates with both constructs (Drake et al., 2008; Zhang & Howell, 2011). Besides Past Positive and Past Negative, it is not clear in which ways Present Hedonistic, Present Fatalistic and Future are related with LS, since findings are inconclusive. However, it results important to note that several studies suggest that individuals who exhibit a Balanced Time Perspective (BTP) display higher levels of LS and mindfulness (Olivera-Figueroa et al., 2016).

Balanced Time Perspective

Balanced Time Perspective (BTP) is considered an ideal and healthier temporal frame, and is characterized by low scores on Past Negative and Present Fatalistic, moderate scores on Present Hedonistic and high scores on Past Positive and Future (Stolarski et al., 2015). In this regard, individuals characterized by BTP would display a flexible switching between individuals’ past, present, and future time orientations (Boniwell & Zimbardo, 2004). Research on BTP has found this TP profile to be associated with higher positive affect, well-being, self-determination, vitality, gratitude (Zhang et al., 2013), positive mood (Stolarski et al., 2013), emotional intelligence (Stolarski et al., 2011), mindfulness (Drake et al., 2008; Seema & Sircova, 2013), and increased cortisol secretion during exposure to stressful experiences (Olivera-Figueroa et al., 2015). Furthermore, BTP has also been demonstrated to serve as a protective factor against psychopathology (Oyanadel & Buela-Casal, 2014). As such, BTP appears to represent a TP construct reflective of an optimal mix of TPs.

Mindfulness’s Relationship to Time Perspective

Ongoing debates exist with respect to the association of mindfulness with TPs (Seema & Sircova, 2013). Some of these debates posit that mindfulness could be: 1) an antecedent of specific TPs (Seema & Sircova, 2013); 2) a process in and of itself (independent of TP); and/or 3) a variable that induces healthy relations toward time, most notably, a BTP. Regarding the view of mindfulness being an antecedent of TP, mindfulness has been considered to represent a TP in and of itself, given its intrinsic temporal orientation implications, resulting from mindfulness’ focus on the present (Seema & Sircova, 2013). On the other hand, the view of mindfulness being a process in of itself suggests that it functions as a mechanism in its own right (i.e.: independent of TP), through the components of (a) self-regulation of attention; and (b) orientation to experience (Bishop et al., 2004). On the view of mindfulness being a variable that induces healthy relations toward time, one review addressing the role of TP on well-being, LS, and mindfulness also synthesized the findings of research studies which suggest that mindfulness can be a variable that induces BTP (Olivera-Figueroa et al., 2016). While bearing in mind that the described debates on mindfulness’ relationship with TP remain ongoing, in the present study, mindfulness appears conceptualized and treated as a TP, due to the framework of the present study being centered around TP theory.

Cross-Cultural Relationships Between Time Perspective, Mindfulness, and Life Satisfaction

The underlying influence of complex cultural paradigms is often manifested in differences between countries regarding cognitions, behaviors, and normative practices (Peng & Nisbett, 1999). With the presentation of divergent views and cognitions between countries, different systems of cultural and psychological co-construction are being shown. On this basis, as well as on the known significant variability between Asian, European and North American cultures, we conducted the current study.

Due to the relevance that TP, mindfulness and LS have for mental health, several studies have addressed relationships between these constructs across many diverse nations throughout the world (Olivera-Figueroa et al., 2016). In terms of comparative studies, it results important to note that only one study has thus far addressed cross-cultural comparisons of TP and LS between countries (Boniwell et al., 2010). Similarly, only one study has addressed cross-cultural interrelationships between TP, mindfulness and LS between countries (Stolarski et al., 2016).

In the study examining the relationship between TP and LS cross-culturally, data was collected in Great Britain and Russia, and subsequently compared through a “person-oriented” hierarchical cluster analysis (HCA) conducted on the TP from both samples (Boniwell et al., 2010). The clusters compared from the HCA steaming of both samples were: (1) hedonistic, present-oriented; (2) future-oriented; (3) BTP; (4) fatalistic, or Negative Time Perspective (NTP); and (5) risk-taking. Results on the British sample revealed that higher LS was associated with positive TP patterns (BTP, hedonistic and future-oriented), whereas lower LS was associated with fatalistic, or NTP patterns. On the Russian sample, higher LS appeared to be associated with BTP and hedonistic patterns, whereas lower LS appeared to be associated with Negative TP patterns (NTP and risk-taking). As such, both British and Russian samples reflected similar relationships between LS and the BTP, NTP and hedonistic patterns.

In the study examining the interrelationships between TP, LS, and mindfulness, the authors tested the potential role of BTP in the relationship between LS and mindfulness across three samples, two of which were conducted on Polish participants, and one comprising both Dutch and German participants (Stolarski et al., 2016). Correlation analysis revealed that LS was negatively associated to Deviation from Balanced Time Perspective (DBTP), yet positively related to mindfulness. Hierarchical regressions revealed that BTP and mindfulness shared a considerable portion of variance (5–12%), thereby proving their incremental validity in predicting LS. Finally, mediation analyses revealed partial mediation, where the total effect of mindfulness on life satisfaction was significantly reduced upon the inclusion of the DBTP across all samples. As such, on this comparative study BTP was found to mediate the relationship between mindfulness and LS of Polish, Dutch and German individuals.

Objectives

As the reviewed comparative studies indicate (Boniwell et al., 2010; Stolarski et al., 2016), associations between TP, LS and mindfulness has indeed been demonstrated in European countries. However, no previous study has thus far examined the cross-cultural role of TP and mindfulness on LS across countries steaming from different continents (Olivera-Figueroa et al., 2016). To address this gap, the present study aimed to assess the role of TP and mindfulness on LS across four countries spanning three continents (North America, Asia, and Europe): the USA, Spain, Poland, and Japan. The countries of the USA, Poland and Japan were selected due to a prior study having reported similarities and differences on well-being across these countries (Kohn et al., 1990). Moreover, recent studies had specifically addressed relationships between LS and temporality-relevant constructs like perceived changes, gratitude, and self-construals in the USA and Japan (Akutsu et al., 2016; Hong et al., 2019; Robustelli & Whisman, 2018). Furthermore, LS had also recently been comparatively studied in relation to temporality across the USA, Japan, and Spain (de Jonge et al., 2017), hence the inclusion of Spain on the present study.

Derivation of Hypotheses

Boniwell and colleagues showed in the examination of a British sample that higher scorings on LS were associated with positive TPs, whereas lower scorings on LS were associated with Past Negative or Present Fatalistic TP (Boniwell et al., 2010). Similarly, in the Russian sample of the same study, higher scorings on LS were linked to Present Hedonistic, whereas lower scorings on LS were associated with Past Negative. Furthermore, in their 2016 study, Stolarski and colleagues found a positive linkage between LS and mindfulness (Stolarski et al., 2016). Therefore, our first hypothesis is:

H₁: LS will be positively associated with mindfulness and Past Positive and negatively associated with Past Negative across the investigated cultures.

Drake and colleagues found in their study conducted in a sample of two hundred and sixty Scottish participants that happiness and mindfulness were positively correlated across individuals characterized by BTP (Drake et al., 2008). Moreover, Boniwell and colleagues found in the Russian sample of their study, that higher scorings on LS were linked to BTP (Boniwell et al., 2010). Additionally, a study by Seema and Sircova conducted in a sample of 610 Estonian students revealed that individuals characterized by BTP exhibited higher scores on mindfulness (Seema & Sircova, 2013). This finding was supported by a study conducted by Rönnlund and colleagues in a Swedish sample, which showed that a higher level of mindfulness promotes a more balanced time perspective (Rönnlund et al., 2019). Therefore, we hypothesized that:

H₂: A mindful disposition and the Deviation from a Negative Time Perspective (DNTP) will predict higher LS across the investigated cultures.

In line with Stolarski and colleagues (Stolarski et al., 2016), who found that LS was negatively associated to Deviation from Balanced Time Perspective (DBTP), we hypothesized that:

H_3: The Deviation from a Balanced Time Perspective (DBTP) will predict decreased LS across the investigated cultures.

Methodology

Participants

Table 1 provides socio-demographic information on the participants of this study. Psychology undergraduate students participated in this cross-cultural study (N = 867, M_AGE = 20.19, SD = 3.417). Of these 867 students, 270 were men (M_AGE = 19.65, SD = 4.275) and 597 were women (M_AGE = 20.43, SD = 2.923). Participants were recruited by promoting the study in various courses, as well as by personal recruitment at the campuses of the involved institutions. All subjects signed informed consent forms and filled out the questionnaires in a paper-pencil manner. In the USA the participants were students from the University of Northern Colorado-School of Psychological Sciences (n = 125, M_AGE = 18.84, SD = 6.995). Japanese participants were students from the University of Tsukuba-Graduate School of Comprehensive Human Sciences in Tsukuba (n = 220, M_AGE = 20.95, SD = 0.732). In Spain the participants were students from the Universitat Autònoma de Barcelona-Department of Health and Clinical Psychology in Barcelona (n = 303, M_AGE = 19.46, SD = 2.438) and in Poland the participants were students from the University of Warsaw-Faculty of Psychology (n = 219, M_AGE = 21.22, SD = 2.513).

Table 1.

Mean scores and standard deviations of age, ZTPI, DBTP, and DNTP coefficients, MAAS and SWLS by countries

	Total (n =867)		Spain (n =303)		Poland (n =219)		USA (n =125)		Japan (n =220)
	females=597 males=270		females=214 males=89		females=160 males=59		females=60 males=65		females=163 males=57
	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Age	20.19	3.42	19.46	2.44	21.22	2.51	18.84	6.99	20.95	0.73
ZTPI
Past Negative (n=4)	2.71	0.91	2.37	0.83	2.77	1.05	3.02	0.61	2.97	0.85
Past Positive (n=3)	3.35	0.81	3.20	0.90	3.33	0.83	3.55	0.69	3.49	0.68
Present Hedonistic (n=3)	3.05	0.87	3.00	1.01	3.15	0.90	3.05	0.56	3.06	0.77
Present Fatalistic n(n=3)	2.97	0.94	3.61	0.74	2.32	0.86	3.26	0.59	2.58	0.77
Future (n=4)	3.44	0.80	3.41	0.85	3.56	0.83	3.62	0.58	3.25	0.77
DBTP	2.90	0.83	3.31	0.77	2.64	0.88	2.78	0.57	2.67	0.76
DNTP	2.61	0.82	2.83	0.73	2.79	0.93	2.17	0.62	2.39	0.76
MAAS (n=15)	3.56	0.85	3.58	0.76	3.96	0.75	3.90	0.78	2.96	0.76
SWLS (n=5)	4.46	1.17	4.53	1.08	4.39	1.10	5.09	1.17	4.07	1.20

Notes: ZTPI=Zimbardo Time Perspective Inventory; DBTP = Deviation Balanced Time Perspective coefficient; DNTP = Deviation from Negative Time Perspective coefficient; MAAS=Mindfulness Attention Awareness Scale; SWLS= Satisfaction with Life Scale

Ethical Statement and Informed Consent

All procedures performed in our study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Our treatment of all samples complied with the ethical standards of the American Psychological Association (APA) Code of Ethics (American Psychological Association, 2017). Moreover, proper approval to conduct the study was obtained from the research ethics boards of the University of Northern Colorado-School of Psychological Sciences, the University of Warsaw-Faculty of Psychology, the Universitat Autònoma de Barcelona-Department of Health and Clinical Psychology, and the University of Tsukuba-Graduate School of Comprehensive Human Sciences. Furthermore, informed consent was obtained from the student participants of the study at the aforementioned universities.

Outcome Measures

Zimbardo Time Perspective Inventory

The Zimbardo Time Perspective Inventory (ZTPI) was developed to measure TP across five tendencies: Past Positive; Past Negative; Present Hedonistic; Present Fatalistic; and Future (Zimbardo & Boyd, 1999). The original questionnaire is composed of 56 items that assess different aspects of all five TP tendencies. Each item follows a 5-point Likert scale format, ranging from very characteristic (5) to very uncharacteristic (1). However, we have used the 17-item version of the scale (Orosz et al., 2017), since the psychometrics of the 56 version are problematic with respect to Confirmatory Factor Analysis (CFA) model fit (Anagnostopoulos & Griva, 2012; Sircova et al., 2014; Worrell & Mello, 2007). Regarding the 17-item version of the ZTPI (Orosz et al., 2017), the psychometric properties reported for its validation demonstrated adequate internal consistency for Past Positive (α = .68, n = 3), Past Negative (α = .84, n = 4), Present Hedonistic (α = .73, n = 3), Present Fatalistic (α = .69, n = 3), and Future (α = .70, n = 4). For this study we administered the 17 items identified on the Orosz et al. 2017 article from the original English language version validated in the USA (Zimbardo & Boyd, 1999), the Spanish language validated version in Spain (Díaz-Morales, 2006), the Polish language validated version in Poland (Kozak & Mazewski, 2007) and the Japanese language validated version in Japan (Shimojima et al., 2012). An example item for Past Negative is “I think about the bad things that happened to me in the past”, whereas the following serves as an example item for Past Positive: “Happy memories of good times spring readily to mind”. A typical item for Present Hedonistic is “I take risks to put excitement in my life” and for Present Fatalistic: “My life is controlled by forces I cannot influence”. The Future TP is measured by items like “I complete projects on time by making steady progress”. In order to ensure configural measurement invariance, we conducted pairwise CFAs for the ZTPI-17. It was shown that the factorial structure was similar across the four countries. The goodness of model fit of most pairwise CFAs were not optimal, but acceptable for most pairwise comparisons (ZTPI-17: all IFI, TLI and CFI > .843). Only the comparisons between Poland vs. USA (IFI = .835, TLI = .787, CFI = .828) and the one between USA vs. Japan (IFI = .818, TLI = .763 CFI = .808) showed overall too low values. We assume that the low observed fit indices are caused by the lower sample-size of the US-sample of n = 125. For reasons of transparency, we report the whole range of fit indices for all pairwise comparisons. Spain-Poland: IFI = .911, TLI = .885, CFI = .909; Spain-USA: IFI = .882, TLI = .849, CFI = .878; Spain-Japan: IFI = .916, TLI = .892, CFI = .913, and Poland-Japan: IFI = .879, TLI = .844, CFI = .875. The term “acceptable” can be backed up by several authors, who question the strict application of the threshold value of >.95 for IFI, TLI and CFI (Barrett, 2007; Marsh et al., 2004; McIntosh, 2007; Nye & Drasgow, 2011). Niemand and Mai (2018, p. 1153) suggest: “Rather than applying a universal threshold (e.g., a value close to .95 for CFI), cutoff points for complex models with small samples should be more ‘forgiving’ (CFI below .95), while they should be stricter for simple models with large samples (CFI above .97)” (Niemand & Mai, 2018).

We tested the metric and scalar invariance of ZTPI-17, as well as for the MAAS and the SWLS, by applying the CMIN-Diff test between the unconstrained model and those with constrained weights (for testing metric invariance) and the constrained intercepts (for testing scalar invariance). For the ZTPI-17 the CMIN-Diff test revealed that the measurement weights model showed a significant difference to the unconstrained model (CMIN-Diff = 330,496, df = 36, p <.001) and that the intercepts model showed a significant difference to the unconstrained model (CMIN-Diff = 1841,546, df = 87, p <.001). Thus, we were not able to observe metric and scalar invariance for the ZTPI-17. For further investigating the possibility of a partial metric invariance of the used ZTPI scale version, we conducted a comparison between an unconstrained model and several constrained models across all four country groups. The purpose of this procedure is to identify critical items in order to improve the cross-cultural studying of time perspectives in future research. In this re-analysis we did not test for scalar invariance. In a first step we observed no full metric measurement invariance by comparing the unconstrained with the constrained model; CMIN-Diff = 28.643, df = 16, p = .004. In the next step we tried to identify if one of the five time perspective dimensions is responsible for the observed metric measurement invariance. For this purpose we created five new models. In each of them we let one of the five dimensions unconstrained whereas all other corresponding four dimensions remained constrained. The comparisons with the unconstrained model revealed that the following time perspective dimensions were not the main source of the missing metric invariance: Present Hedonistic (CMIN-Diff = 23.583; df = 13, p = .009), Past Positive (CMIN-Diff = 25.256, df = 13, p = .005.), Future (CMIN-Diff = 24.988, df = 12, p = .003.), and Present Fatalistic (CMIN-Diff = 26.235, df = 13, p = .003.). In contrast the comparison between the unconstrained model and the model with unconstrained Past Negative items (but with all other items being constrained), revealed that this dimension can be assumed to be the main source of the lack of metric invariance (CMIN-Diff = 44.951, df = 12, p = .092).

In a third step we intended to further narrow down the critical items. For this analysis we included the four items of the Past Negative, as well as four other items which show striking differences of factor loading across the four country groups, indicated by very low loading in at least one of the involved country groups. By creating several model combinations of these suspicious items we intended to find an insignificant comparison with the unconstrained model, which has the lowest number of critical items. As a result we observed a model with three critical items, two of them from the Past Negative dimension and one item from the Present Hedonistic dimension (CMIN-Diff = 17.220, df = 13, p = .070). We can conclude that these three items should be further scrutinized in future studies to ensure measurement invariance of the used 17-ZTPI version in cross-cultural research. The identified three items appear below, with the numeration in parentheses referring to the one in the original 56-item version of the ZTPI by Zimbardo and Boyd (1999) as displayed by Orosz et al. (2017, p. 6): a) It’s hard for me to forget unpleasant images of my youth (Item 34, Past Negative), b) I think about the bad things that have happened to me in the past (Item 50, Past Negative) and c.) I find myself getting swept up in the excitement of the moment (Item 46, Present Hedonistic). This can imply reformulations or replacement of these items, as well as theoretical consideration about possible different biased understanding in the involved cultures.

Despite the individuality that each TP tendency entails, subsequent TP research has identified a distinct TP profile known as Balanced Time Perspective (BTP). Individuals displaying low scores in Past Negative and Present Fatalistic, moderate scores in Present Hedonistic and high scores in Past Positive and Future are considered to reflect the BTP profile. To measure the BTP profile one coefficient has been developed: the Deviation from the Balanced Time Perspective (DBTP) coefficient (Stolarski et al., 2011). This coefficient examines how individuals deviate from this profile. Because DBTP measures the distance of an individual from an optimal TP profile, the farther a DBTP value is from zero, the more misbalanced an individual’s TP profile is considered to be. The formula to calculate DBTP is as follows:

$$DBTP=\sqrt{{(oPN-ePN)}^2+{(oPP-ePP)}^2+{(oPF-ePF)}^2+{(oPH-ePH)}^2+{(oF-eF)}^2}$$

In this formula o equals the observed value obtained for each measured TP, whereas e equals the expected optimal value for each TP, as indicated by Zimbardo and Boyd [1.95 for Past Negative (PN), 4.6 for Past Positive (PP), 1.5 for Present Fatalistic (PF), 3.9 for Present Hedonistic (PH), and 4.0 for Future (F)] (Zimbardo & Boyd, 2008).

For this study, we specifically chose to analyze BTP in our data through the DBTP coefficient because it has previously exhibited better predictive validity with regard to the well-being/LS of individuals than other BTP operationalization methods (Zhang et al., 2013), such as the cut-off point approach (Drake et al., 2008), and cluster analysis (Boniwell et al., 2010). Thus, we feel that through our use of the DBTP coefficient in our study we have addressed cultural validity, since the DBTP coefficient has been validated across several countries in North America, Europe and Asia (Stolarski et al., 2020). As such, it seems that the DBTP coefficient’s previous implementation in North American, European and Asian countries provide the most approximate BTP optimal/suboptimal indices that we could have utilized to measure BTP in our study conducted across four countries spanning the aforementioned three continents (North America, Asia, and Europe).

Furthermore, we have also calculated the Deviation from a Negative Time Perspective (DNTP) coefficient, which was developed by Oyanadel and Buela-Casal in 2014, as an alternative for measuring the distance of an individual from a maladaptive TP profile (Oyanadel & Buela-Casal, 2014). The DNTP is calculated as follows:

$$DNTP=\sqrt{{(nPN-ePN)}^2+{(nPP-ePP)}^2+{(nPF-ePF)}^2+{(nPH-ePH)}^2+{(nF-eF)}^2}$$

In this formula n equals the observed negative value obtained for each measured TP, whereas e equals the expected negative value for each TP, as indicated by Zimbardo, Sword and Sword [4.35 for Past Negative (PN), 2.80 for Past Positive (PP), 3.30 for Present Fatalistic (PF), 2.65 for Present Hedonistic (PH), and 2.75 for Future (F)] (Zimbardo et al., 2012).

Mindful Attention Awareness Scale

The trait Mindful Attention Awareness Scale (MAAS) is a 15-item scale designed to assess mindfulness as a receptive state of mind, in which attention simply observes what is taking place, in a manner informed by sensitive awareness of what is occurring in the present (Brown & Ryan, 2003). Through many studies conducted across undergraduate, community and nationally sampled adult and adult cancer populations since the scale’s introduction in 2003, the trait MAAS has consistently shown excellent psychometric properties (Carlson & Brown, 2005; MacKillop & Anderson, 2007). For instance, its internal consistency levels (Cronbach’s alphas) usually range from .80 to .90. Moreover, across studies the trait MAAS has continually confirmed a single factor scale structure. Furthermore, the MAAS has also demonstrated high test-retest reliability, discriminant and convergent validity, known-groups validity, and criterion validity. Overall, the trait MAAS usually takes about 5 minutes or less to complete. The internal consistency for the MAAS was high (α = .86, n = 15). An example item of the MAAS is “I find it difficult to stay focused on what’s happening in the present.” To address cultural validity of the MAAS in our study, we administered the original English language version in the USA (Brown & Ryan, 2003), the Spanish language version in Spain (Soler et al., 2012), the Polish language version in Poland (Jankowski, 2014), and the Japanese language version in Japan (Wakamatsu et al., 2011). Configural measurement invariance was addressed in this study by conducting pairwise CFAs for the MAAS. The goodness of model fits of all pairwise CFAs were sufficiently high (MAAS: all IFI, TLI and CFI > .900).

For the difference between the unconstrained model and the measurement weights model, we observed that the difference was not significant on the MAAS (CMIN-Diff = 74,31, df = 42, p = .150). However, the corresponding difference between the unconstrained model and the intercepts modes reached significance (CMIN-Diff = 1148,622, df = 87, p < .001). Thus, we observed metric but no scalar measurement invariance for the MAAS across the four countries.

Satisfaction with Life Scale

The Satisfaction with Life Scale (SWLS) is a short, 5-item instrument that measures global cognitive judgments of satisfaction with one’s life, as a whole (Diener et al., 1985). This scale exhibits good convergent validity with other scales, and with other assessments of subjective well-being. Life satisfaction, as assessed by the SWLS, exhibits a degree of temporal stability (e.g., .54 for 4 years). Moreover, the scale displays discriminant validity from emotional well-being measures. The SWLS usually takes 1 minute or less to complete. The internal consistency for the SWLS appeared to be good (α = .81, n = 5). An example item of the SWLS is “If I could live my life over, I would change almost nothing”. To address cultural validity of the SWLS in our study, we administered the original English language version in the USA (Diener et al., 1985), the Spanish language version in Spain (Atienza et al., 2000; Vazquez et al., 2013), the Polish language version in Poland (Jankowski, 2015), and the Japanese language version in Japan (Sumino, 1994). Measurement invariance was addressed by conducting pairwise CFAs for the SWLS. Pairwise CFAs for the SWLS showed that the goodness of model fits of all pairwise CFAs was sufficiently high (SWLS: all IFI, TLI and CFI > .900).

For the SWLS we did not observe significance for the metric invariance (CMIN-Diff = 17,825, df = 12, p = 12.12), but we found a significant difference for the scalar invariance (CMIN-Diff= 172,001, df = 27, p <.001). Thus again, we observed metric but no scalar measurement invariance for the SWLS. To summarize we demonstrated configural invariance for all three used scales. Further MAAS and LS showed metric invariance and the ZTPI-17 scale showed partial metric invariance. The lack of full metric measurement invariance does not necessarily indicate a problem with the ZTPI-17 scale. Instead, there may be a different interpretation of the concepts in different cultures, which needs to be further addressed in future studies.

Statistical Analyses

In term of statistics, for this study correlation analyses were conducted for age, all TP sub-scales, the DBTP and DNTP coefficients, the MAAS and the SWLS across all participants of this cross-cultural study. Then, one-way ANOVAs were conducted to assess LS, mindfulness, TPs, DBTP and DNTP as a function of culture. Subsequently, hierarchical linear regression analysis (HLRA) were conducted on data from each country with LS as the outcome (dependent) variable. Moreover, mediation analyses were also conducted to address whether the effect of mindfulness on LS is mediated by DBTP and DNTP in the USA, Spain, Poland and Japan, as well as across all samples analyzed together. Finally, a reverse mediation analysis was conducted to consider reversed relations, in the way that mindfulness mediates the relationship between DBTP, DNTP and LS. All analyses were conducted with SPSS-26 and Amos-27. Furthermore, for the mediation analyses we applied the PROCESS Macro-3.3 (Hayes, 2019). Regarding missing data, there were only few missing data. These data were replaced by mean-values of the corresponding variables.

Results

Table 2 reports correlations for age, all TP sub-scales, the DBTP and DNTP coefficients, the MAAS and the SWLS across all participants of this cross-cultural study. Table 2 reports additionally descriptive statistics and correlations for the same variables across the USA, Spain, Poland and Japan. We included all pairs of variables with a significant correlation in at least one country.

Table 2.

Pearson’s correlations among age, TP subscales, DBTP, DNTP, MAAS and SWLS in USA, Spain, Poland and Japan

		Spain	Poland	USA	Japan
Age	PN	.140 *	−.029	−.040	−.021
Age	PP	.029	−.142 *	.058	.014
Age	DNTP	−.143 *	−.016	.078	.035
Age	MAAS	−.163 **	.130	.083	.010
Age	SWLS	−.126 *	−.019	.048	−.018
PN	PP	−.140 *	.037	−.095	−.318 **
PN	PF	.149 **	.247 **	.105	.157 *
PN	DBTP	.353 **	.332 **	.519 **	.586 **
PN	DNTP	−.802 **	−.723 **	−.625 **	−.767 **
PN	MAAS	−.205 **	−.204 **	−.227 *	.270 **
PN	SWLS	−.245 **	−.343 **	−.308 **	−.425 **
PP	PH	−.019	.313 **	.014	.219 **
PP	PF	−.004	.128	.235 **	.042
PP	F	.116 *	.139 *	.239 **	.059
PP	DBTP	−.555 **	−.606 **	−.325 **	−.568 **
PP	DNTP	.258 **	.066	.575 **	.479 **
PP	MAAS	.216 **	−.041	.086	−.078
PP	SWLS	.433 **	.053	.189 *	.262 **
PH	PF	−.024	.219 **	.125	.161 *
PH	F	−.092	−.021	.262 **	−.055
PH	DBTP	−.342 **	−.489 **	−.310 **	−.353 **
PH	DNTP	.226 **	−.483 **	.135	.173 *
PF	DBTP	.616 **	.179 **	.522 **	.457 **
PF	DNTP	−.096	−.558 **	.044	−.317 **
PF	MAAS	−.110	−.196 **	−.164	.233 **
PF	SWLS	−.101	.038	−.146	−.178 **
F	DBTP	−.265 **	−.309 **	−.364 **	−.274 **
F	DNTP	.250 **	.209 **	.468 **	.262 **
F	MAAS	.329 **	.159 *	.209 *	−.215 **
F	SWLS	.245 **	.195 **	.117	.081
DBTP	DNTP	−.293 **	−.169 **	−.429 **	−.569 **
DBTP	MAAS	−.228 **	−.099 **	−.251 **	.272 **
DBTP	SWLS	−.327 **	−.203 **	−.321 **	−.359 **
DNTP	MAAS	.308 **	.231 **	.181 *	−.280 **
DNTP	SWLS	.364 **	.193 **	.238 **	.408 **
MAAS	SWLS	.440 **	.217 **	.313 **	−.289 **

ZTPI = Zimbardo Time Perspective Inventory; PN = Past Negative; PP = Past Positive; PH = Present Hedonistic; PF = Present Fatalistic; F = Future; DBTP = Deviation from Balanced Time Perspective; DNTP = Deviation from Negative Time Perspective; MAAS = Mindfulness Awareness Attention Scale; SWLS = Satisfaction with Life Scale

^*p < .05

^**p < .01

significant correlations are printed in bold types

Subsequently, one-way ANOVAs were used to assess LS, mindfulness, TPs, DBTP and DNTP as a function of culture. These analyses revealed a significant difference between countries in: LS (F (_{3, 863}) = 22.27, η²_p = .072, p < .0001); mindfulness (F (_{3, 863}) = 74.46, η²_p = .206, p < .0001); Past Negative (F (_{3, 863}) = 27.64, η²_p = .088, p < .0001); Past Positive (F (_{3, 863}) = 8.23, η²_p = .028, p < .0001); Future (F (_{3, 863}) = 8.13, η²_p = .028, p < .0001); Present Fatalistic (F (_{3, 863}) = 150.31, η²_p = .348, p = < .0001); DBTP (F (_{3, 863}) = 44.38, η²_p = .134, p < .0001), and DNTP (F (_{3, 863}) = 31.48, η²_p = .099, p < .0001), as well as a lack of significant difference between cultures in Present Hedonistic (F (_{3, 863}) = 1.30, p = .274). To analyze the main effect of culture, pairwise comparisons were performed using a Bonferroni correction for multiple comparisons. These analyses revealed that Spanish (M = 4.53, SD = 1.08) and Polish Europeans (M = 4.39, SD = 1.10) exhibited less LS than Americans (M = 5.09, SD = 1.17), but greater LS than Japanese (M = 4.07, SD = 1.19), all p’s < .0001 (J<P<S<USA). With respect to mindfulness, Spanish Europeans (M = 3.57, SD = 0.75) exhibited less mindfulness than Americans (M = 3.90, SD = 0.78), but greater mindfulness than Japanese (M = 2.96, SD = 0.75). Polish Europeans (M = 3.96, SD = 0.75) displayed more mindfulness than Spaniards, Americans and Japanese, all p’s < .0001 (J<S<USA<P). Americans (M = 3.02, SD = 0.61) scored higher on Past Negative than Spaniards (M = 2.37, SD = 0.83), Polish (M = 2.77, SD = 1.05), and Japanese (M = 2.97, SD = 0.85), all p’s < .0001 (S<P<J<USA). Regarding Past Positive, Spanish (M = 3.20, SD = 0.90) and Polish Europeans (M = 3.32, SD = 0.83) exhibited less Past Positive than Americans (M = 3.55, SD = 0.69) and Japanese (M = 3.49, SD = 0.68), all p’s < .0001 (S<P<J<USA). In terms of Present Hedonistic, the difference in the means of the Spanish (M = 2.99, SD = 1.01), Polish (M = 3.15, SD = 0.90), American (M = 3.05, SD = 0.56), and Japanese samples (M = 3.06, SD = 0.77) did not reach significance, all (p = .274). Regarding Present Fatalistic, Americans (M = 3.26, SD = 0.59) exhibited less Present Fatalistic than Spaniards (M = 3.61, SD = .74), but greater Present Fatalistic than Polish Europeans (M = 2.31, SD = 0.86) and Japanese (M = 2.58, SD = 0.77), all p’s < .0001 (P<J<USA<S). In relation to Future, Spanish (M = 3.41, SD = 0.85) and Polish Europeans (M = 3.56, SD = 0.83) exhibited lower scores than Americans (M = 3.62, SD = 0.57) and higher scores than Japanese (M = 3.25, SD = 0.77), all p’s < .0001 (J<S<P<USA). Concerning the DBTP, Polish Europeans (M = 2.63, SD = 0.88) exhibited less DBTP than Japanese (M = 2.66, SD = 0.76), Americans (M = 2.77, SD = 0.57), and Spaniard Europeans (M = 3.30, SD = 0.77), all p’s < .0001 (P<J<USA<S). Last but not least, with respect to the DNTP, Americans (M = 2.17, SD = 0.62) displayed less DNTP than Japanese (M = 2.39, SD = 0.76), Polish (M = 2.79, SD = 0.93) and Spanish Europeans (M = 2.83, SD = 0.73), all p’s < .0001 (USA< J<P <S).

In light of the detected difference in our dependent variable of LS across cultures, a hierarchical linear regression analysis (HLRA) was conducted on data from each country with LS as the outcome (dependent) variable. The first block included age and sex as controlling variables, utilizing the enter method. The second block included all TPs as predictors, utilizing the stepwise method. The third block included the DBTP coefficient as a predictor, utilizing the enter method. The fourth block included the DNTP coefficient as a predictor, utilizing the enter method. The fifth block included mindfulness as a predictor, utilizing the enter method. The described five block approach was implemented to measure the additive predictive value of each variable assessed through the third, fourth and fifth blocks, beyond the predictive value of the controlling variables assessed through the first and second blocks. Following the described methods, one HLRA was conducted for each of the four countries.

Main Analyses (per country)

Spaniards

Multiple regression analyses conducted on the sample of Spaniards yielded seven models (see Table 3). Of these seven models, the seventh and final one provided the best model fit (R square change = .064; p < .001), as well as the largest R and R square values (R = .596; R square: .355). Past Positive predicted increased LS (β = .302, B = .364, SE B = .073, t = 4.977, p < .0001). In addition, mindfulness also predicted increased LS (β = .282, B = .405, SE B = .075, t = 5.395, p < .0001).

Table 3.

Hierarchical linear regression with Age, Gender, ZTPI subscales, DBTP, DNTP and MAAS as predictors of SWLS scores by countries

	Spain
	β	Sig. β	R	R2	ΔR2	ΔF	Sig. ΔF
Model 7			.596	.355	.064	29.103	<.0001
Age	−.059	.223
Gender	.070	.141
Past Negative	.028	.776
Past Positive	.302	<.0001
Present Hedonistic
Present Fatalistic
Future	.053	.377
DBTP	−.037	.578
DNTP	.187	.053
MAAS	.282	<.0001
	Poland
	β	Sig. β	R	R2	ΔR2	ΔF	Sig. ΔF
Model 6			.448	.201	.028	7.260	.008
Age	−.062	.331
Gender	.031	.627
Past Negative	−.462	<.0001
Past Positive
Present Hedonistic
Present Fatalistic
Future	.215	.002
DBTP	.011	.877
DNTP	−.217	.026
MAAS	.175	.008
	USA
	β	Sig. β	R	R2	ΔR2	ΔF	Sig. ΔF
Model 6			.449	.202	.042	6.227	.014
Age	.032	.713
Gender	.095	.281
Past Negative	−.219	.096
Past Positive	.165	.159
Present Hedonistic
Present Fatalistic
Future
DBTP	−.146	.165
DNTP	−.091	.525
MAAS	.217	.014
	Japan
	β	Sig. β	R	R2	ΔR2	ΔF	Sig. ΔF
Model 7			.493	.243	.024	6.752	.010
Age	−.026	.666
Gender	−.029	.636
Past Negative	−.276	.014
Past Positive	.130	.159
Present Hedonistic
Present Fatalistic	−.077	.364
Future
DBTP	−.012	.916
DNTP	.055	.628
MAAS	−.166	.010

Notes: ZTPI=Zimbardo Time Perspective Inventory, DBTP=Deviation Balanced Time Perspective coefficient; DNTP=Deviation Negative Time Perspective coefficient; MAAS=Mindfulness Attention Awareness Scale; SWLS=Satisfaction with Life Scale.

Polish

Multiple regression analyses conducted on the sample of Polish Europeans yielded six models (see Table 3). Of these six models, the sixth and final one provided the best model fit (R square change = .028; p = .008), as well as the largest R and R square values (R = .448; R square: .201). While Future (β = .215, B = .284, SE B = .092, t = 3.086, p = .002) and mindfulness predicted increased LS (β = .175, B = .256, SE B = .095, t = 2.695, p = .008), Past Negative (β = −.462, B = −.479, SE B = .103, t = −4.649, p < .0001) and DNTP predicted decreased LS (β = −.217, B = −.255, SE B = .114, t = −2.248, p = .026).

Americans

Multiple regression analyses conducted on the sample of Americans yielded six models (see Table 3). Of these six models, the sixth model provided the best model fit (R square change = .042; p = .014), as well as the largest R and R square values (R = .449; R square: .202). In this model only mindfulness predicted increased LS (β = .217, B = .324, SE B = .130, t = 2.495, p = .014).

Japanese

Multiple regression analyses conducted on the sample of Japanese participants yielded seven models (see Table 3). Of these seven models, the seventh one provided the best model fit (R square change = .024, p = .010), as well as the largest R and R square values (R = .493, R square = .243). Past Negative predicted decreased LS (β = −.276, B = −.389, SE B = .158, t = −2.468, p = .014). An unexpected result was that mindfulness scores predicted decreased LS (β = −.166, B = −.264, SE B = .101, t = −2.598, p = .010).

Mediation Analyses

In order to elaborate more deeply on the relations between the constructs, mediation analyses were conducted, to address whether the effect of mindfulness on LS is mediated by DBTP and DNTP in the USA, Spain, Poland, and Japan, as well as across all samples analyzed together. The DBTP showed mediation effects between mindfulness and LS in two out of four samples.

The indirect effect through DBTP in Spain was .027 (LLCI) and .107 (ULCI) with 5,000 Bootstrap samples (indirect effect = .062, BC95%CI from .027 to .107). In Poland there was no indirect mediating effect of the DBTP (indirect effect = .023, BC95%CI from −.011 to .068). The same holds true for Japan, where there was no indirect mediating effect of the DBTP (indirect effect = −.068, BC95%CI from −.156 to .012). In contrast, the USA, there was an indirect mediating effect of the DBTP (indirect effect = .081, BC95%CI from .005 to .194). Thus, the DBTP partially mediates the relationship between mindfulness and LS in Spain and in the USA, but not in Poland and Japan. Moreover, across all samples analyzed together, the BC95% C.I. for the indirect effect through DBTP was −.005 (LLCI) and .030 (ULCI) with 5,000 Bootstrap samples (indirect effect = .0123, BC95%CI from −.005 to .030).

Regarding the DNTP, it showed mediation effects between mindfulness and LS in three out of four samples. In Spain the BC95% C.I. for the indirect effect through DNTP was .040 (LLCI) and .150 (ULCI) with 5,000 Bootstrap samples (indirect effect = .090, BC95%CI from .040 to .150). In Poland the BC95% C.I. for the indirect effect through DNTP was .001 (LLCI) and .105 (ULCI) with 5,000 Bootstrap samples (indirect effect = .042, BC95%CI from .001 to .105). In Japan the BC95% C.I. for the indirect effect through DNTP was −.234 and −.039 with 5,000 Bootstrap samples (indirect effect = −.119, BC95%CI from −.234 to −.039). Thus, there was an opposite mediating effect in Japan. In the USA there was no indirect mediating effect of the DNTP (indirect effect = .027, BC95%CI from −.027 to .103). The described mediation analyses revealed that in our data DNTP partially mediates the relationship between mindfulness and LS in Spain, Poland and Japan (reversed), but not in the USA (see Fig. 1). Across all samples analyzed together, the BC95% C.I. for the indirect effect through DNTP was .019 (LLCI) and .063 (ULCI) with 5,000 Bootstrap samples (indirect effect = .039, BC95%CI from .019 to .063).

Figure 1:

Parallel Multiple Mediator Analyses on DBTP’s and DNTP’s role on the influence of mindfulness on life satisfaction (LS).

Relationships between DBTP, mindfulness and life satisfaction from 4 samples. A*B represents indirect effect, C’ is the direct and C is the total effect of mindfulness on life satisfaction.

*p < .05; **p < .001; ***p < .0001 (numbers accompanying letters refer to studies 1, 2, 3, and 4, and all countries addressed on the study).

1. Spain

2. Poland

3. USA

4. Japan

5. All countries combined

Reverse Mediation Analysis

To consider reversed relations, in the way that mindfulness mediates the relationship between DBTP, DNTP and LS, a reverse mediation analysis was conducted. The aim was to show whether the effect of DBTP and DNTP on LS is mediated by mindfulness in the USA, Spain, Poland, and Japan, as well as across all samples analyzed together. Results from these analyses showed that mindfulness mediates the relationship between DBTP and LS in three out of four samples. In Spain the BC95% C.I. for the indirect effect through mindfulness was −.190 (LLCI) and −.065 (ULCI) with 5,000 Bootstrap samples (indirect effect = −.122, BC95%CI from −.190 to −.065). In Poland there was no mediating effect (indirect effect = −.024, BC95%CI from −.074 to .011). In Japan, the BC95% C.I. for the indirect effect through mindfulness was −.179 (LLCI) and −.021 (ULCI) with 5,000 Bootstrap samples (indirect effect = −.088, BC95%CI from −.179 to −.021). In the USA the BC95% C.I. for the indirect effect through mindfulness was −.278 (LLCI) and −.014 (ULCI) with 5,000 Bootstrap samples (indirect effect = −.126, BC95%CI from −.278 to −.014). In Spain, Japan and USA life satisfaction is being reduced through DBTP when mediated by mindfulness. Across all samples analyzed together, the BC95% C.I. for the indirect effect through mindfulness was not significant (indirect effect = −.015, BC95%CI from −.038 to .006).

Moreover, mindfulness mediates the relationship between DNTP and LS in three out of four samples. In Spain the BC95% C.I. for the indirect effect through mindfulness was .099 (LLCI) and .240 (ULCI) with 5,000 Bootstrap samples (indirect effect = .165, BC95%CI from .099 to .243). In Poland the BC95% C.I. for the indirect effect through mindfulness was .010 (LLCI) and .097 (ULCI) with 5,000 Bootstrap samples (indirect effect = .050, BC95%CI from .010 to .097). In Japan the BC95% C.I. for the indirect effect through mindfulness was .019 (LLCI) and .163 (ULCI) with 5,000 Bootstrap samples (indirect effect = .083, BC95%CI from .019 to .163). In the USA there was no indirect mediating effect of mindfulness (indirect effect = .096, BC95%CI from −.002 to .238). The described reverse mediation analyses revealed that in our data, mindfulness does partially mediate the relationship between DNTP and LS in Spain, Poland, and Japan (see Fig. 2). Across all samples analyzed together, the BC95% C.I. for the indirect effect through mindfulness was .020 and .067 with 5,000 Bootstrap samples (indirect effect = .042, BC95%CI from .020 to .067).

Figure 2:

Reverse mediation analysis on mindfulness’s role on the influence of DBTP and DNTP on life satisfaction (LS).

Relationships between mindfulness, DBTP and DNTP and life satisfaction from 4 samples. A*B represents indirect effect, C’ is the direct and C is the total effect of mindfulness on life satisfaction.

*p < .05; **p < .001; ***p < .0001 (numbers accompanying letters refer to studies 1, 2, 3, and 4, and all countries addressed on the study).

1. Spain

2. Poland

3. USA

4. Japan

5. All countries combined

Discussion

The present study addressed the role of TP, BTP and mindfulness on LS across the USA, Spain, Poland and Japan. First of all, it is worth noting that students from the United States showed the highest scores in LS. The described result is supported by the findings of a previous study on positions in the class structure and psychological functioning of employed men in the United States, Japan, and Poland, which identified employers, managers, and first-line supervisors in the United States to have the strongest sense of well-being, as well as be the least distressed, when compared to their counterparts in Japan and Poland (Kohn et al., 1990). Because psychology university students typically pursue degrees in order to eventually attain work roles such as employers, managers, and first-line supervisors, it makes sense for our intercontinental study to have yielded the same pattern of findings on well-being/LS as those reported by Kohn and colleagues (Kohn et al., 1990) in the United States, Japan, and Poland.

Besides the described similarity in findings between the present study and the study by Kohn and colleagues (Kohn et al., 1990), we also acknowledge that our finding of students from the United States having showed the highest scores in LS contrasts results from previous studies suggesting that North Americans rank lower than Europeans in LS (Alesina et al., 2004; Blanchflower & Oswald, 2004; Di Tella et al., 2003). This difference could be explained by the present study’s focus on psychology students, since previous studies suggest that sample characteristics, study designs and other psychosocial factors can influence LS levels (Alesina et al., 2004) thereby reinforcing LS’ variability across nations (Diener et al., 2010).

In terms of our study’s results, the HLRAs conducted on each assessed country yielded several statistically significant findings. First, in our study Past Negative predicted decreased LS across two of the four countries addressed in this study (Poland and Japan). Therefore, our first hypothesis about LS being expected to yield negative associations to Past Negative across the investigated cultures, was supported across our data from Poland and Japan, but not in those from Spain and the USA. In line with some of our findings, previous studies also found that Past Negative negatively correlated with LS in Poland (Maciuszek et al., 2019), in the United Kingdom (Drake et al., 2008), and in the Asian country of Taiwan (Gao, 2011). Because Past Negative tends to correlate with depression (Zimbardo & Boyd, 1999), the lack of statistical significance in Past Negative predicting decreased LS in Spain in our study could be explained by the findings of a study where prevalence of Major Depressive Episode (MDE) was found to be lower in Spain than in other Western countries (Gabilondo et al., 2010). Similarly, the lack of statistical significance in Past Negative predicting decreased LS in the USA in our study could be explained by the findings of a study where over two million American adults appeared to misperceive feeling depressed from having clinical depression (Cao et al., 2020).

Second, while Past Positive significantly predicted increased LS in Spain, it did not do so in the USA, Poland, nor in Japan. As such, our first hypothesis about LS being expected to yield positive associations to Past Positive was only fully supported across our data from Spain. This finding of Spain being the only country where Past Positive fully predicted LS can be explained by the findings of a study that reported a high degree of perceived LS across Catalan populations (Batista-Foguet et al., 1996), since the data collected in Spain for this study was collected in a sample of Catalan students.

Third, in our study mindfulness predicted increased LS in the European samples of Spain and Poland, a result supported by a previous study, where mindfulness disposition measured through the MAAS appeared to be related to LS in Great Britain (Brown et al., 2009). Similarly, mindfulness predicted increased LS in the USA sample, supporting findings from studies conducted in the USA with the MAAS, where mindfulness was also found to be positively associated to LS (Brown & Kasser, 2005; Brown et al., 2009; Brown & Ryan, 2003). In contrast, mindfulness surprisingly appeared to decrease LS in Japan. The discrepancies in mindfulness detected across the cultures investigated on this study can be explained by previous research findings suggesting that Eastern and Western conceptualizations of mindfulness may have important differences (Christopher et al., 2009). For instance, the MAAS was developed in a Western culture (USA) on the basis of Caucasian-American, psychology-based conceptualizations of mindfulness. As such, the American-based MAAS may not generalize to Eastern cultures’ conceptualizations of mindfulness that are more directly influenced by Buddhist faith (Christopher et al., 2009). Thus, our second hypothesis’ statement about LS being expected to yield positive associations to mindfulness was only supported across our data from the USA, Spain and Poland.

Fourth, the DNTP predicted decreased LS in Poland. Thus, our second hypothesis about LS being expected to yield positive associations to DNTP was not supported. The finding of DNTP having predicted decreased LS in Poland could be explained by the findings of a study where DNTP was positively correlated to increased cortisol systemic output (Olivera-Figueroa et al., 2015), which is a physiological response characteristic of increased stress.

Fifth, in contrast to our third hypothesis, the HLRAs revealed that the DBTP did not predict decreased LS across any of the cultures investigated in this study. Moreover, the finding that the DBTP has a mediating role on mindfulness and LS in our Spanish and American samples is in line with the findings of a previous study, where DBTP was found to mediate the relationship between mindfulness and LS across Polish, Dutch and German individuals (Stolarski et al., 2016). Moreover, our findings suggest that the DNTP is a mediator of the relationship between mindfulness and LS in Spain, Poland and Japan (in an opposite direction on the latter). Thus, our third hypothesis about LS being expected to yield negative associations to DBTP was supported across two of the cultures investigated in this study. The discrepancies in the results between our study and previous studies could be attributed to the use of different approaches to the ZTPI measurement across these studies.

Regarding the lack of statistically significant findings in this study across the Present scales, we need to note that several studies addressing relationships between TP and LS though the ZTPI and SWLS did not report relationships between the Present scales and LS. In the case of Present Hedonistic, studies conducted in Belgium (Desmyter & De Raedt, 2012) and Portugal (Ortuño et al., 2015) did not detect statistically significant relationships between Present Hedonistic and LS. Similarly, studies conducted in Belgium (Desmyter & De Raedt, 2012), the Netherlands (Vowinckel, 2012) and Portugal (Ortuño et al., 2015) also did not found statistically significant relationships between Present Fatalistic and LS. However, it results important to note that studies conducted in Netherlands (Vowinckel, 2012), France (Gana et al., 2013), Taiwan (Gao, 2011), the USA (Zhang & Howell, 2011), Great Britain and Russia (Boniwell et al., 2010), did identify strong correlations between Present Hedonistic and LS. Moreover, the latter three studies additionally identified a relationship between Present Fatalistic and LS.

In the case of Future, studies conducted in Belgium (Desmyter & De Raedt, 2012), the Netherlands (Vowinckel, 2012), France (Gana et al., 2013), and Portugal (Ortuño et al., 2015) failed to detect statistically significant relationships between this TP and LS. In contrast, the Future reached significance in studies from Great Britain, Russia (Boniwell et al., 2010), Taiwan (Gao, 2011), and the USA (Zhang & Howell, 2011). In our study, Future reached significance in relation to LS only in Poland. Importantly, this finding of Future being a positive predictor of LS was also reported in another study conducted across Polish students (Maciuszek et al., 2019).

While some of the findings of the cited studies were similar to ours, there also exist notable differences, which could be traced back to the different research designs applied. For instance, the study of Vowinckel employed a Dutch version of the ZTPI, which may have triggered more positive associations on the Present Hedonistic scale due to differences in connotation structures that are language-inherent (Vowinckel, 2012). The study by Gana and colleagues (Gana et al., 2013) consisted of a cross-sectional study to address the mediating effects of self-efficacy on the relationship between TP and LS, while controlling for age, sex, self-perceived health, and depressive mood, of which the latter two variables were not addressed in this study. The study by Boniwell and colleagues (Boniwell et al., 2010) measured LS in Great Britain through the Temporal-Satisfaction with Life Scale (TSWLS) sub-scales (Pavot et al., 1998), which is a different LS measurement than the one utilized in this study. The study by Gao employed a logistic regression analysis to discriminate between high and low satisfaction groups, which found that low Past Negative scores and high Past Positive or Future scores were more likely to be assigned to the high LS group than to the low LS group (Gao, 2011), which is an approach not utilized on the present study. The study by Zhang and Howell studied whether TP predicts unique variance in LS beyond personality traits (Zhang & Howell, 2011), which is a variable not addressed in the present study.

The important overall finding related to the differences found in the constructs between the four countries can be explained by: 1) The different operationalizations of mindfulness across cultures, 2) additional factors related to LS, and 3) cultural differences in TPs in between different societies. While in the cultures of origin mindfulness is part of a larger system of Buddhist belief and practice with strong ethical and moral dimensions, for the Western cultures mindfulness is understood as a mode of present-centered. Examining mindfulness in other than the originated social contexts may change the nature and effects of the practice (Kirmayer, 2015). In addition, LS has been linked in studies to wealthy nations, and thus to economic factors (Vittersø et al., 2002). As for TPs, it could be the case that some societies are more hedonistic oriented than others, in which discipline and future orientation are parts of socialization.

Limitations and future directions

In terms of limitations, the present study was cross-sectional, only utilized one questionnaire to assess each construct of interest, and the sample was composed of only psychology undergraduate students. Therefore, future studies should be driven by longitudinal designs to disentangle the extent to which mindfulness practice might regulate BTP and LS. In terms of cultural validity, future studies should include more countries within the continents addressed on this cross-cultural study. Doing so would allow for more reliable and robust conclusions on the cross-cultural similarities and differences of TP, mindfulness and LS. Future studies should also recruit a more representative sample with a wider socio-demographic range, including different academic and professional characteristics.

Although we have used validated scale versions of the SWLS and the MAAS, as well as validated items of the ZTPI in all four countries, we observed some non-invariances at the metric and scalar level in the case of the ZTPI-17. Consequently, there is a need to improve the ZTPI measurement instrument, as the use of different versions leads to some differences in results. While configural invariance was shown, the observed fit indices were not optimal, but were at least acceptable. Nonetheless, we are aware that the particularly non-optimal fit indices are an important limitation of our study and thus we will draw two conclusions: 1) the importance of a cautious interpretation of our results, in particular those referring to the comparisons including the USA, and 2) the need of future studies, which can highly benefit from the results of the current study. Moreover, we inspected the factor loading of all conducted pairwise comparisons to deliver important information about the precise shortcomings of the ZTPI-subscales. The following items showed very low loadings: two Past Negative items (items 22 and 25 in the original ZTPI-56-version) for Spain; one Present Hedonistic item (item 46) for Japan and one Past Positive item for Japan and Poland (item 15). Furthermore, the reported items with low factor loadings indicate the lack of metric invariance, which demonstrate the necessity to improve short-versions of the ZTPI, in particular for its use in cross-cultural research by future studies. Future research is also needed to further modify the applied measurement scales. We observed that several items are responsible in the current study for not reaching the full metric measurement invariance for the ZTPI-17. We recommend to test possible modification of those items in order to enable a further improvement of the application of the used ZTPI scale version in the context of cross-cultural studies.

Nonetheless, our results have made a valuable contribution, since configural and partly scalar and metric invariance was shown. It is however important to interpret the validity of our results cautiously against the background of the need to improve the measurement instruments. This is essential for cross-cultural studies in general and should be addressed in particular for the ZTPI-17.

Similarly, future research should address the influence of the BTP (measured through DNTP) on the LS in various countries, as well as the question of whether BTP promotes a mindful disposition or if the mindful practice promotes BTP. Furthermore, future studies should also address the role that LS plays in the effect that mindfulness-based interventions (MBIs) play on promoting BTP (Rönnlund et al., 2019). Longitudinal studies are also warranted to answer these questions. Moreover, future studies should re-examine the role of Past Negative in the reduction of LS, as well as the role of Past Positive in the increase of LS. Finally, the effects of mindfulness on LS in Japan should be investigated in further detail, in order to address an understanding of the cultural context from the regions around which the mindfulness concept originated (Christopher et al., 2009).

Another limitation of the present study is that the MAAS is only one out of about ten mindfulness-measures validated across adult populations (Park et al., 2013). Moreover, the MAAS solely focuses on the attention component of mindfulness and does not measure other aspects of the construct, e.g., non-judging or curiosity. Thus, the findings of our study could be interpreted as reflective of the receptive state of mindfulness measured by the MAAS, as well as the trait nature of mindfulness, rather than mindfulness practice. As such, a follow-up study using different mindfulness questionnaires is warranted. Furthermore, future studies could address other aspects of mindfulness like non-judging and/or curiosity, as well as relationships of both trait mindfulness and mindfulness practice with TP and LS across cultures and continents. A final general limitation is that the current study relied solely on self-reported data. Future studies could benefit from other measurement approaches such as behavioral tasks.

Clinical Implications

Despite its methodological limitations, the present study suggests that cognitive interventions could address TP regulation by focusing on positive past experiences, as well as by reducing the impact of negative memories (Zimbardo et al., 2012). Similarly, the present study also suggests that MBIs could be included to help balance dysfunctional TPs at least in Western societies (Rönnlund et al., 2019), as well as increase LS, since mindfulness promotes well-being and overall mental health (Gu et al., 2015). Moreover, through the described mindful practice techniques individuals could also be assisted on developing mindfulness dispositional traits to be present in everyday life (Brown & Ryan, 2003).

Conclusions and Contributions

The present study represents the first attempt to analyze the cross-cultural role of TP and mindfulness on LS across the USA, Spain, Poland and Japan. Our findings support the notion that a mindful disposition is a good predictor of LS at least in Western societies. Additionally, the DNTP was found to mediate the relationship between mindfulness and LS in three countries with the exception of the USA. Overall, this study presents a relevant contribution to the field of subjective well-being, since it shows for the first time that the association of TP, mindfulness and LS differs across the investigated countries as a function of culture.

Data availability statement

From this study, part of the data collected in Spain was included in this published article: Muro, A., Feliu-Soler, A., Castellà, J., Deví, J., & Soler, J. (2017). Does time perspective predict life satisfaction? A study including mindfulness as a measure of time experience in a sample of Catalan students. Mindfulness, 8(3), 655–663. Similarly, part of the data collected in Poland was included in this published article: Stolarski, M., Vowinckel, J., Jankowski, K. S., & Zajenkowski, M. (2016). Mind the balance, be contented: Balanced time perspective mediates the relationship between mindfulness and life satisfaction. Personality and Individual Differences, 93, 27–31.