Future Time Perspective and Real-Life Utterances About the Future in Young and Older Adults
Abstract
Abstract. Future time perspective (FTP) refers to an individual’s global perception of the future. It has been found to be positively related to life satisfaction. FTP is traditionally assessed via self-report, but recently a few studies have used observable behaviors for assessing FTP. We focused on two real-life behaviors (frequency and qualities of talking about the personal future) and explored whether they could be used as behavior-based measures of FTP. We examined the association between these behaviors and self-reported FTP, and their relationships with life satisfaction. The sample included 55 young (aged 18–31) and 47 older adults (aged 62–83) who completed questionnaires on future time perspective and life satisfaction. Over 4 days, participants carried an electronically activated recorder, which randomly captured 30-second sound snippets from their daily lives – a total of 30,656 sound snippets were collected. Participants’ utterances were coded for temporal orientation. Linguistic inquiry word count was used to analyze the qualities of future-oriented utterances. Structural equation models showed that self-reported FTP was not associated with the two real-life behaviors. It was positively associated with life satisfaction for the whole sample. The frequency of future-oriented utterances and family-related words were positively related to young adults’ life satisfaction. Achievement-related words were positively related to older adults’ life satisfaction.
The view of life as having a beginning and an end is inevitably rooted in the concept of time. Recently, growing interest has emerged for understanding how an individual’s perception of time influences their lives. Time perspective is broadly defined as the construct underlying an individual’s use of the past, present, or future to organize the continual flow of their experiences (Zimbardo & Boyd, 1999). It has been described as one of the most influential determinants of human behavior (Boniwell & Zimbardo, 2004). With regards to understanding the effects of time perspective on behavior and how this may be influenced by age, particular interest lies in perspectives related to the future, given that it sets the stage upon which an individual may yet act. Future time perspective (FTP) represents an individual’s perceptions of the future and their remaining time to live (Coudin & Lima, 2011; Rohr, John, Fung, & Lang, 2017). Empirical evidence has emphasized the importance of this construct for shaping an individual’s well-being, motivation, and behavior (e.g., Demiray & Bluck, 2014; Kooij, Kanfer, Betts, & Rudolph, 2018) as well as its relationships to processes of aging (e.g., Cate & John, 2007; Lang & Carstensen, 2002).
To date, the construct of FTP has been investigated almost exclusively based on traditional self-report methods (e.g., Lu, Li, Fung, Rothermund, & Lang, 2018). Recently, opportunities offered by modern technology have inspired a few researchers to explore novel approaches to predicting FTP from observable behavior (e.g., social media posts; Schwartz et al., 2015). The present study is the first to focus on the observable behavior of talking about the personal future: Using a naturalistic observation method in everyday life, namely, the Electronically Activated Recorder (EAR; Mehl, Pennebaker, Crow, Dabbs, & Price, 2001), we recorded random snippets of daily conversations of young and older adults over 4 days. We coded how much (i.e., frequency) and how (i.e., qualities) young and older adults talked about their personal future in daily life. The first goal of the study was to explore whether these two behaviors could be used as behavior-based measures of FTP. Thus, we examined whether they were associated with FTP as measured by the traditional self-report measure (i.e., Future Time Perspective Scale). We explored whether individuals’ subjective and global perception of their future was associated with how much and how they talked about their future. The second goal of the study was to examine the widely studied relationship between FTP and life satisfaction (e.g., Seijts, 1998; Simons, Vansteenkiste, Lens, & Lacante, 2004; Zimbardo & Boyd, 1999). We also explored whether our two behavioral measures would show the same relationship with life satisfaction. The final goal was to examine all of these associations in young versus older adults, as young adults tend to have a more positive and open-ended FTP than older adults (e.g., Demiray & Bluck, 2014). In sum, we examined whether individuals’ global perspective of their future reflects onto their daily language use and how these relate to their life satisfaction.
Future Time Perspective and Well-Being in the Context of Aging
One of the most widely used conceptualizations of FTP is that proposed by Carstensen and Lang (1996), who described it as a bipolar dimension extending from perceiving the future as limited to open-ended. Although there has been a recent shift toward a multidimensional operationalization of FTP (for an overview, see Rohr et al., 2017), Carstensen and Lang’s (1996) single bipolar dimension remains the most commonly used conceptualization (Brothers, Chui, & Diehl, 2014; Rohr et al., 2017). Rohr and colleagues (2017) proposed that the Future Time Perspective Scale can be reliably implemented for assessing FTP both as a single construct as well as a multidimensional one, depending on the focus of the research question at hand.
Regarding FTP, one of the most consistently reported interactions is that with chronological age (e.g., Brothers et al., 2014; Cate & John, 2007; Brothers, Gabrian, Wahl, & Diehl, 2016; Coudin & Lima, 2011; Demiray & Bluck, 2014; Grühn, Sharifian, & Chu, 2016; Lang & Carstensen, 2002). Age accounts for the time from birth till the present, whereas FTP reflects views on the perceived time remaining between the present and the end of life. Thus, the inevitable process of aging is assumed to have inverse effects on these two constructs.
FTP has been associated with psychosocial development (Kruger, Reischl, & Zimmerman, 2008), as well as the formation of personal goals over the lifetime (Carstensen, Isaacowitz & Charles, 1999). For example, the socioemotional selectivity theory (Carstensen et al., 1999) portrays perception of the future and of one’s time left to live as a crucial factor for the age-related shift of prioritizing instrumental goals during young adulthood and then emotion regulation ones in later life. According to this theory, selecting goals congruent to one’s perceived time left to live plays a key role in determining an individual’s well-being and, thus, the natural shifts in FTP present an adaptive mechanism within the context of aging (Lang & Carstensen, 2002).
Nevertheless, Coudin and Lima (2011) oppose this conclusion. They found the relationships between goals typical of open-ended FTP and well-being to be positive for adults of all ages, with these benefits being especially pronounced for individuals with limited FTP. Given that limited FTP was also associated with older age, they suggested that open-ended FTP may have positive effects on well-being for individuals of all ages. In fact, a multitude of studies have provided evidence supporting the notion that expanded views of the future are related to positive life outcomes related to well-being (Kooij et al., 2018; Seijts, 1998; Simons et al., 2004; Zimbardo & Boyd, 1999). More specifically, open-ended FTP has been positively related to subjective well-being (Allemand, Hill, Ghaemmaghami, & Martin, 2012; Coudin & Lima, 2011), psychological well-being (Demiray & Bluck, 2014), satisfaction with life (Brothers et al., 2016; Park, et al., 2015; Schwartz et al., 2015), and positive affect (Grühn et al., 2016). On the other hand, negative relationships have been found between open-ended FTP and negative affect (Allemand et al., 2012; Hicks, Trent, Davis, & King, 2012), and depressive symptoms (Grühn et al., 2016).
Considering the relationships between FTP and aspects of well-being within the context of aging, the negative association between FTP and age would imply that age-related decreases in FTP are coupled with decreases in well-being. In contrast though, evidence depicts a curvilinear trend for well-being, with it being highest in older age (Ramsey & Gentzler, 2014). Regarding this incongruence, the fact that most literature on FTP has relied on self-report methods may play a role. In response to this limitation, recently researchers have started to investigate FTP independently of self-reports and to explore behavior-based methods (Park et al., 2015; Schwartz et al., 2015).
Novel Approaches to FTP Research: Behavioral Data
Most existing studies assessed FTP based solely on self-report, which comes with a number of limitations. Apart from the classical limitations associated with self-report (i.e., impression management, self-deceptive enhancement, participant awareness, memory biases; Mehl, Robbins & Deters, 2012), experts highlight a few specific weaknesses related to assessing FTP. First, there is a considerable overlap between self-reported FTP and personality trait Conscientiousness (Dunkel & Weber, 2010; Adams & Nettle, 2009; Webley & Nyhus, 2006; Zimbardo & Boyd, 1999), which complicates the distinction between these two constructs (Park et al., 2015; Schwartz et al., 2015). Second, self-reported scales themselves may inherently contribute to age-biases depending on the specific wording of certain items (Brothers et al., 2014). For example, an item such as “Most of my life lies ahead of me” (FTPS; Carstensen & Lang, 1996) might generate different responses from individuals in their twenties compared to those in their seventies, simply because of the objective reality of their age, independent of their actual FTP.
Recently, researchers began to explore novel approaches to gathering data independently of self-report. Developments in smartphone technology and the advent of the Internet have provided new means for gathering behavioral data in real life settings (e.g., Miller, 2012). Some of these approaches (e.g., mobile sensing; Harari, Müller, Aung, & Rentfrow, 2017) enable collecting data independently of self-report and present three major advantages: They produce objective data, which can be reliably and meaningfully quantified; they are characterized by high ecological validity having been collected in the real world; and given their unobtrusive nature, they extend data collection to subtle and unconscious behaviors otherwise inaccessible to self-report methods (Mehl et al., 2012). One such method is the electronically activated recorder (EAR), a recorder that can be programmed to capture snippets of ambient sounds in real time. The EAR is used to collect objective data of auditory nature, such as one of humans’ most powerful social behaviors, namely, real-life language (Manson & Robbins, 2017).
Natural language, in the form of everyday conversations, has often been overlooked by researchers given its mundane and apparently unimportant appearance (Duck & Usera, 2014). A number of studies, however, revealed that language is a relevant source of behavioral data, which can then be used to predict individual characteristics, such as age, sex, personality (Boyd & Pennebaker, 2017; Ireland & Mehl, 2014; Park et al., 2015; Pennebaker & Stone, 2003; Schwartz et al., 2013), and, recently, time perspective (Park et al., 2015; Schwartz et al., 2015). Park and colleagues (2015) used written linguistic data in the form of Twitter and Facebook postings to assess an individual’s time perspective (i.e., past, present, future) and their relationship to age, personality, life satisfaction, and depression. Based on the temporal orientation ratings of three independent judges, they developed an automated temporal classification model. Findings showed open-ended FTP to be positively related to age and life satisfaction, and negatively to depression. The authors concluded that their innovative methodological approach largely supported and replicated the literature on time perspective, suggesting that behavioral data are appropriate means for investigating this construct.
Although Park and colleagues’ (2015) novel method and findings are exciting, two limitations of the study are worth addressing. First, they did not use any other temporal orientation measures (e.g., self-report) to ensure that their novel method measures the same underlying construct as existing measures. Hence, it is unclear whether their future orientation measure assesses FTP. This might explain why they found open-ended FTP to be positively related to age, which is inconsistent with the negative relationship reported in the FTP literature (e.g., Lang & Carstensen, 2002). This finding could also be related to the narrow age range of their sample (i.e., 13–48 years), given that FTP becomes predominantly associated with limitations after the age of 60 (Strough, de Bruin, Parker, Lemaster, Pichayayothin, & Delaney, 2016). Second, although data from social network sites are easily accessible and abundant, according to Walther (2007), computer-based communication “differs substantially from face-to-face communication, in form if not in function” (p. 2539) and that the “time spent in computer-based communication prompts especially mindful and deliberative message composition” (p. 2543).
The Current Study
Similar to Park and colleagues’ (2015) approach, we investigated FTP in relation to a language-based, objective behavior in real-life settings. However, our study had three advantages over Park and colleagues’ design (2015): We included a self-report measure of FTP; we included not only young individuals, but also older adults in our sample; and we focused on naturalistic conversations rather than computer-based language. We examined language in the form of everyday utterances using a naturalistic observation method, the EAR. Seijts (1998) claimed that perceiving the future as expansive leads individuals to be more involved in the future, to think more about it, and to act more upon it. Thus, theoretically, FTP might be consistently traceable in an individual’s natural language use. Everyday conversations with others are one of our prominent means for interacting with the social world in which we live and develop as well as playing a role in maintaining our self-concepts (Pasupathi, Mansfield, & Weeks, 2014). The relationship between FTP and daily talking behavior has not (yet) been investigated, and this relationship could provide a deeper insight into another way in which FTP affects our daily life.
The first goal of the study was to explore whether the frequency and qualities of future-oriented utterances could be used as behavior-based measures of FTP. Thus, we examined whether they were associated with FTP as measured by the traditional self-report measure (i.e., Future Time Perspective Scale). The second goal, based on the literature reporting a positive relationship between open-ended FTP and life satisfaction, was to investigate the relationships of the three measures with life satisfaction. Finally, the third goal of the study was to consider these within the context of aging and compare relationships for young and older adults. All investigated associations are depicted in our conceptual model in Figure 1.
We assessed a naturally occurring, future-oriented behavior, namely, that of talking about one’s future in daily life. We investigated two aspects of future-oriented utterances: a quantitative measure of frequency and a qualitative one reflecting how people talk about their future. Regarding the former, we expected the frequency of future-oriented utterances to be positively related to self-reported FTP. Based on the negative relationship between age and FTP (e.g., Grühn et al., 2016; Lang & Carstensen, 2002), we expected that young adults would talk more frequently about their personal future compared to older adults. On the other hand, because open-ended FTP is positively associated with life satisfaction (e.g., Park et al., 2015), regardless of age (e.g., Demiray & Bluck, 2014), we expected positive relationships between the frequency of future-oriented utterances and life satisfaction for both young and older adults.
Considering our qualitative measure, we used the Linguistic Inquiry and Word Count (LIWC; Meier, Boyd, Pennebaker, Mehl, Martin, Wolf, & Horn, 2018; Pennebaker & Francis, 1999) to analyze the patterns of specific word category use in future-oriented utterances. Based on the socioemotional selectivity theory (Carstensen et al., 1999), we focused on qualities of future-oriented utterances related to emotions (i.e., use of positive and negative emotion words), close social relationships (i.e., words related to family and friends), and drives (i.e., words related to affiliation and achievement). Although there is a vast literature on how individuals perceive their future (Lang & Damm, 2017), to date no study has investigated how people talk about their future in everyday life. Thus, we openly explored the relationships of the six qualities of future-oriented utterances for young and older adults. We explored whether, for example, there was a positive relationship between the number of positive emotion words in future-oriented utterances and life satisfaction (in contrast to a negative relationship between the number of negative emotion words in future-oriented utterances and life satisfaction).
Methods
Participants
This study is part of a larger project, which recruited and collected real-life sound snippets of 111 healthy (Swiss) German-speaking young and older participants. Participants were recruited through the participant pool of the Department of Psychology at the University of Zurich, via flyers or through advertisements on websites and in local newspapers. An inclusion criterion for older adults was a minimum score of 27 on the Mini Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975). Nine participants were excluded due to inadequate data: Seven had missing questionnaires, and two spoke predominantly a foreign language. Thus, the final sample, described in Table 1, was composed of 102 adults, split into two age groups: young adults (18–31 years) and older adults (62–83 years).
Procedure
Introductory Session
Before the 4 days of audio recording started, participants met with the research team and received instructions on the study, as well as completing a package of questionnaires including measures of demographics and future time perspective. They also received their assigned iPhone and charging cable. Participants were informed that the iPhones were set to “airplane mode” and locked with only the EAR application running, thus functioning solely as a recorder that randomly and beyond their awareness would record 30-s snippets of their daily lives. They were instructed to carry the iPhone with them as much as possible over the next 4 days.
The EAR
We used the iPhone app “EAR 2.0” (Mehl, 2014). The EAR was programmed to automatically record 30-s snippets of daily life at random times (on average four times per hour) between 6 am and midnight over the span of 4 days (two weekdays and one weekend, counterbalanced). Participants were instructed to carry the iPhone with them as much as possible during the day, either attached to a belt or in their pocket, and to charge it every night. Participants were also asked to fill out daily end-of-the-day diaries for the 4 days reporting a general overview of their main activities (e.g., eating, studying, spending time with friends) and when each activity occurred.
Feedback Session
After the 4 days of recording, participants again met with the research team and completed further questionnaires including one assessing their life satisfaction. They also evaluated the EAR method on a 10-item questionnaire. Documentation of method acceptance and compliance, as well as dropout rate is available at https://osf.io/yd6qu/. Participants were also given the opportunity to listen and delete any of their sound files that they desired to exclude from the study. Finally, they received a CD containing all of their sound files and were compensated with either 50 Swiss Francs or, in the case of university students, with course credit.
Measures
Self-report measures were administered before (T1) and after (T2) audio data collection. For this study, we investigated future time perspective assessed at T1 and life satisfaction at T2 in line with the order in our conceptual model.
Future Time Perspective
Participants’ subjective perception of their future was assessed using the German version of Carstensen and Lang’s Future Time Perspective Scale (FTPS; 1996). This scale is composed of 10 items each rated on a 7-point Likert scale. FTPS had a high reliability, Cronbach’s α = .91.
Life Satisfaction
Life satisfaction was assessed using the Satisfaction with Life Scale (SWLS; Diener, Emmons, Larsen & Griffin, 1985; German translation from Schumacher, 2003). The five items of the SWLS were rated on a scale from 1 (do not agree at all) to 7 (fully agree) and had an internal consistency of Cronbach’s α = .83.
Behavioral Measures
All sound files were listened to and coded (by two coders) for whether the participant was talking or not. Files containing participant’s speech were transcribed and coded for temporal orientation (mean interrater reliability: 85%). To protect the privacy of others, transcription and coding focused solely on participants’ utterances, without including those of other social partners. Temporal orientation coding included a reference to time (i.e., talking about the past, present, and/or future) and to person (i.e., talking about one’s self or others; see Demiray, Mehl, & Martin, 2018, for a description of the coding scheme). Personal future referred to anything that will/might or will/might not happen in one’s future (e.g., “We will not go to the movies tomorrow,” “I have to work on Saturday”). All coding categories were dichotomous, indicating presence (1) or absence (0) of a temporal orientation.
Frequency of Future-Oriented Utterances
The frequency of future-oriented utterances was calculated for each participant by dividing the number of their personal future-oriented utterances by the total sum of their temporally oriented utterances (i.e., utterances about both personal and others’ past, present, and future).
Qualities of Future-Oriented Utterances
The qualities of utterances were analyzed using the newest German version of the text analysis program DE-LIWC15 (Meier et al., 2018). This program comprises 18,711 words mapped into over 80 dictionary categories (Meier et al., 2018). Using automated word count analysis, the program permits analyzing use of words related to a specific category (e.g., emotions) from written language or transcribed speech. The qualities of personal future-oriented utterances of each participant were investigated by merging only those sound files containing references to the personal future into a single file and analyzing the frequency of the following six word categories: positive emotions (e.g., happy), negative emotions (e.g., offended), family (e.g., dad), friends (e.g., friendship), affiliation (e.g., relationship), and achievement (e.g., success).
Analytical Approach
The conceptual model is presented in Figure 1 and was analyzed using structural equation modeling (SEM). We used the R package lavaan (Rosseel, 2012) using R Language, version 1.0.136 (RStudio Team, 2016). Confirmatory factor analysis (CFA) was conducted, and modification indices that made theoretical sense within the framework of the model were adopted to fit the model. Given that this study investigated the structure of the relationships between variables across young and older adults, it was not necessary to apply constraints of measurement invariance (otherwise necessary to enable multigroup comparison of variable means; Borsboom, 2006). Model fits were evaluated based on the indices and respective cut-off criteria: chi-square (X2) test statistic with its degrees of freedom and p-value; the Bentler comparative fit index (CFI) considering > .90 and > .95 as cut-offs for “acceptable” and “good” fit, respectively (McDonald & Ho, 2002); the Steiger-Lind root mean square error of approximation (RMSEA) with values close to or less than .08 and a confidence interval of .05 –.10 indicating “acceptable” fit (McDonald & Ho, 2002); and given our small sample size (N < 150), the standardized root mean square residual (SRMR) with a cut-off value close to .08 for “good” fit (Hu & Bentler, 1999). Nevertheless, given the sensitivity and complexity of SEM models, in some cases models with less adequate fits may be retained when these decisions could be justified (Ockey & Choi, 2015) and in the case of our small sample size (N < 150), less rigorous cut-offs are required for retaining all correct models and rejecting incorrect ones (Sivo, Fan, Witta, & Willse, 2006).
Results
Preliminary Analyses
Over the span of 4 days, the EAR captured a total of 30,656 sound files (for further details on sound file deletions, see https://osf.io/yd6qu/). For young adults, 3,442 sound files (18.6% of young adults’ sample) containing speech were collected (M = 62.58, SD = 31.90), whereas for older adults the total was 2,583 files (21.2% of older adults’ sample; M = 54.96, SD = 31.29). An independent samples t-test revealed this difference to be nonsignificant, t(100) = 1.21, p = .23, suggesting that the frequency of talking in general was equal across the two age groups. Subsequent analyses were conducted on only those sound files containing personal future-oriented utterances, which for young adults resulted in a sample of 298 sound files (8.7% of all young participants’ utterances) and 133 (5.1% of all utterances) for older adults.
Table 2 presents an overview of correlations, descriptive statistics and tests of mean differences for young and older adults. After computing the means of the frequency of future-oriented utterances for young and older adults (Table 2), an F-test revealed the two age groups to have unequal variances F(54, 46) = 2.50, p = .002. Thus, we proceeded to using Welch’s two samples t-tests to investigate the mean differences between the two age groups (Table 2). Consistent with the literature on FTP (e.g., Grühn et al., 2016; Lang & Carstensen, 2002) and our expectations, we found that young adults had significantly higher FTP scores than older adults (Table 2). Results showed that, on average, young adults talked significantly more frequently about their future than older adults (Table 2). With regards to life satisfaction, results showed an opposite trend with older adults scoring higher on life satisfaction compared to their younger counterparts, but this difference did not reach significance (Table 2). Finally, while talking about their personal future, young adults mentioned friends more than older adults did (Table 2).
Major Analyses
Given the limitation of our small sample size for testing models with greater numbers of variables, we opted for parsimonious models and tested seven variations of the conceptual model (Figure 1): one including only FTP and frequency of future-oriented utterances (Model 0) and thereafter six models including a different LIWC variable (i.e., words related to positive emotions, negative emotions, family, friends, affiliation, and achievement; Models 1–6). Tables 3 and 4 display SEM regression coefficients for the predictors and the fits for each model, respectively.
In line with our first research goal, we first examined the relationships between self-reported FTP and the two behavior-based measures. We tested whether the frequency of future-oriented utterances was positively related to FTP and examined the covariance between these two variables in Model 0. Although the fit of the model was acceptable (Table 4), FTP and the frequency of future-oriented utterances emerged as unrelated for both young (β = .061, p = .667) and older adults (β = .095, p = .505). For the sake of parsimony, we subsequently dropped this relation from Models 1–6.
Next, we examined whether self-reported FTP and the six qualities of future-oriented utterances (as measured by LIWC) were associated (Table 3). No relationships emerged between self-reported FTP and the qualities of future-oriented utterances, neither for young nor for older adults. Considering the relationships between our two behavior-based measures, only for young adults did we find significant relationships: Words related to positive emotions, affiliation, and achievement were positively related to the frequency of future-oriented utterances.
In line with our second and third research goals, we investigated the associations of FTP and the two behavior-based measures to life satisfaction, and we did so for both young and older adults.1 We found that, although FTP was positively related to life satisfaction in both age groups, the frequency of future-oriented utterances was significantly related to life satisfaction only among young adults (Table 3). With regards to the qualities of future-oriented utterances, only two of the six LIWC variables emerged as related to life satisfaction (Table 3). For young adults, only the number of words related to family was positively associated with life satisfaction (β = .19, p = .01). For older adults, the number of words related to achievement had a marginally significant effect and was positively associated with life satisfaction (β = .23, p = .09).
Finally, we tested an additional model in order to check the robustness of our findings. We tested a single model with only the four predictors that showed significant effects above (i.e., FTP, frequency of future-oriented utterances, Family and Achievement; see Appendix, Table A1). Our results were replicated, however, given the increased complexity of the model with respect to our sample size, the fit was inferior to our original models. The fact that our results were replicated shows the robustness of our original findings.
Discussion
The three goals of the current study were to explore the relationship of the traditional self-report measure of FTP with two real-life behaviors (i.e., frequency and qualities of utterances related to one’s personal future); to examine whether these three measures showed consistent relationships with life satisfaction; and to examine all these associations within the context of aging, considering relationships for young and older adults.
Four main findings emerged from this study. First, self-reported FTP was not related to the frequency or the qualities of future-oriented utterances, neither for young nor for older adults. Second, self-reported FTP was consistently and positively related to life satisfaction for both young and older adults. Third, the frequency of future-oriented utterances was also positively related to life satisfaction, but only for young adults. Finally, age group differences emerged for the qualities of future-oriented utterances, with words related to family being positively associated with young adults’ life satisfaction, whereas words of achievement being relevant for older adults’ life satisfaction.
Considering our first goal of relating FTPS scores with the two real-life behaviors, neither frequency of future-oriented utterances nor any of the six linguistic categories representing qualities of these utterances showed any relationship to FTP. In other words, the frequency with which adults (both young and old) talk about their future and how they talk about it is not related to how they subjectively perceive their future. These findings lead us to question those of Park and colleagues (2015) and the validity of assessing FTP from verbal behavior. Nevertheless, a couple of points are worth noting: Although self-reported FTP and aspects of talking about the future were not directly related to each other, the frequency of future-oriented utterances showed the typical negative relationship with age consistently found with self-reported FTP, supporting the idea of a shared underlying construct. Furthermore, Lang and Damm (2017) emphasize how the use of spatial analogies, such as extension, for representing FTP have notable shortcomings, given that they are based on the assumption that subjective perception adheres to the laws of physical space. Thus, future research is needed to shed more light on the possibility that behavior-based measures may reflect aspects of FTP not captured by self-report and consider the potential of multimethod approaches for studying the same phenomenon from different perspectives.
Another explanation may be that future-oriented thinking might be happening mostly privately (in people’s minds) rather than socially in conversations. Previous experience-sampling studies show that in real life individuals tend to think much more about their future than about their past (i.e., prospective bias; D’Argembeau, Renaud, & Van der Linden, 2011; Felsman, Verduyn, Ayduk, & Kross, 2017; Rasmussen & Berntsen, 2011). In contrast, Demiray and colleagues (Demiray, Mehl & Martin, 2018) found a retrospective bias in real-life conversations, showing that individuals talk about their past two to three times as much as their future. Thus, one’s general view of their future (as measured by the FTPS) may be more likely to be reflected onto their momentary thoughts rather than onto their conversations. Future research should use the experience-sampling method in addition to the FTPS to examine whether individuals with more positive and open-ended FTP also tend to think more frequently about their future (and with different qualities) in everyday life.
The second goal of the study, based on the widely observed positive relationship between FTP and life satisfaction, was to replicate this finding as well as exploring the relationships of the two real-life behaviors with life satisfaction. Consistent with the literature, self-reported FTP was positively related to life satisfaction independent of age (e.g., Demiray & Bluck, 2014). However, the relationship was stronger for older adults than young adults. The reason might be that self-reported FTP was the major predictor of life satisfaction for older adults, but for young adults the frequency of future-oriented utterances was also influential. That is, only for young adults did the frequency of future-oriented utterances mirror the positive relationship of self-reported FTP with life satisfaction: Young adults who talked more about their future were also more satisfied with life, but this relationship did not exist for older adults. Why? Considering self-reported FTP, it is not difficult to imagine why believing that one has longer to live might have positive effects on life satisfaction, as is consistently found in the literature (e.g., Brothers et al., 2016). On the other hand, in older adulthood, the literature describes a human tendency to gradually become more apprehensive about the future (Shmotkin, 1992) as well as exhibiting an increasing strive for a sense of self-continuity through time, resulting in older adults perceiving their future selves as a continuous part of their present selves (Rutt & Löckenhoff, 2016). These tendencies might lead older adults not only to talk less directly about their future, but also to assimilate perceptions and significance of this time frame into the views of their continuous present self. In fact, the literature describes cognitive strategies considered adaptive for successful aging including extending the time frame of one’s present-self and devaluing the future as to emphasize savoring the present (Lang & Damm, 2017). In older age, the relevance of the future may have become integrated into views of the present, which might explain why future-oriented utterances may be less relevant for their life satisfaction. In fact, the present has been found to be positively related to life satisfaction in older age (Lennings, 2000). This is in line with the socioemotional selectivity theory (Carstensen, Isaacowitz & Charles, 1999), which suggests that older adults have more emotion-oriented goals, which can be realized in the present through focusing on positive experiences and reaching achievable goals.
In contrast, young adults who tend to perceive their future as almost limitless in terms of time and opportunities seem to talk more overtly about their future, which is positively associated with their life satisfaction. Young adults emphasize knowledge-oriented goals, focus on gathering as much information about the world as possible and on pursuing activities that will pay off in the future (Carstensen, et al., 1999). In line with these goals, they may be actively planning and making decisions about their activities, career, or relationships, while talking to others in real life (Ciairano, Rabaglietti, Roggero, & Callari, 2010). Young adults of the same culture share (and thus potentially talk about) many common life events they expect to occur in the present or near future, which creates a common conversational ground. By discussing their future plans and goals with others, a young person may gain insightful information or learn from the experiences of others. Thus, overtly talking about the future may give them a sense of control, potentially reduce anxiety, and shine a bright light upon prospects of future success, which in turn could contribute positively to life satisfaction.
Considering the qualities of future-oriented utterances and their relationships with life satisfaction, further age group differences emerged. For young adults, using more words related to family emerged as positively related to life satisfaction, whereas achievement-related words showed significant importance for older adults. One explanation for young adults might be that more references to family reflect stronger familiar relationships, more frequent contact as well as foreseeing the permanence of these relationships in the future. Parental attachment has been positively linked to life satisfaction in young adults (Guarnieri, Smorti, & Tani, 2015). Furthermore, Lambert and his colleagues (2010) found family relationships and their support to be highly important for determining the sense of meaning in life for young adults. Other studies have found positive effects of close family ties on the self-esteem of young adults (e.g., Roberts & Bengtson, 1996). Considering young adulthood as a unique stage of development characterized by a focus on achieving intimacy rather than social isolation (Erikson, 1950), family could provide a secure and meaningful base upon which to rely during the phase of exploration and identity formation, thus contributing to being satisfied with life.
A second and not mutually exclusive explanation could reside in the concept of life scripts (Berntsen & Rubin, 2002; Rubin & Berntsen, 2003). These are culturally shared prototypical life events that contribute to forming expectations about the future as well as the approximate ages at which these events typically occur. Accordingly, several major family-related events fall within the years of young adulthood (e.g., finding a long-term partner, marriage, childbirth). Although it is acknowledged that life scripts do not represent the average life, people tend to consider events occurring “on time” positively (Rubin & Berntsen, 2003), whereas “off-time” occurrences are associated with stigma and stress (Neugarten & Hagestad, 1976). Thus, it would be plausible to assume that for young adults who are approaching or within the age range of these events, fantasizing, being en route, or actually being in the midst of these life events could color their conversations regarding their future, as well as contribute to their life satisfaction.
This point of life scripts finds further support given that family-related words cease to have positive relationships for older adults. No family-related events are to be expected in the age range of our old group. This is somewhat surprising considering that the socioemotional selectivity theory (Carstensen, et al., 1999) describes prioritizing close relationships (i.e., family) to satisfy one’s emotional needs in older age. Findings here indicated that older adults did not refer to family less frequently than young adults, but rather that this frequency was unrelated to their life satisfaction. This may suggest that although older adults highly value the support of their family members, in old age milestones related to family have already been achieved and no longer represent personal future goals toward which to strive for to maintain satisfaction with life.
In contrast, the use of words related to achievement in older adults’ future-oriented utterances was associated with life satisfaction. This is of particular interest given that, although the frequency with which older adults talked about their future was not linked to their life satisfaction, their use of more achievement-related words in these utterances was. Why achievement references bare a particular relevance for older adults might be explained by the nature of aging itself. Havighurst (1972) described dealing with losses and adjusting to changes in one’s physical, cognitive, social, and professional life comprise major developmental tasks of old age. Successful aging entails overcoming these hurdles and is defined by the maintenance of functional capacities and active engagement with life (Rowe & Kahn, 1997). Studies found successful aging to be associated with positive outlooks on life, strong feelings of self-efficacy, sense of control, autonomy, independence as well as effective coping strategies to deal with age-related losses (Bowling & Dieppe, 2005; Erikson, 1950). All of these factors may not only determine older adults’ successful aging in the present, but may also enhance their trust in their abilities, thus enabling envisioning future achievements. From this perspective, talking about future achievements could potentially be an indicator of present successful aging, reflected in greater life satisfaction of older adults who make more references to future achievements.
Limitations and Future Research
One limitation of the present study was the small sample size. Especially when conducting SEMs, the commonly recognized rule of thumb suggests a ratio between sample size and estimated parameters of 20:1 (Kline, 2011). Nevertheless, even with our sample, our SEM models achieved acceptable fits, thus we felt justified interpreting our results. One strength of the study was the big size of our real-life data: We collected thousands of sound snippets from our participants, which provided us with robust data on their everyday language use.
Another limitation was that we did not have a middle-aged group in our study. Based on the documented age-related shift in time perspective occurring during midlife (e.g., Cate & John, 2007), it would also be interesting to observe the frequency of talking about the future and the qualities of future-oriented utterances during this period of life.
We did not find any relationships between self-reported FTP and the two real-life behaviors on talking about the future. Although past research showed that language is a good proxy for indirectly studying FTP, our language-based measures were unrelated to how individuals subjectively perceived their future. Future research should use a multimethod approach with other real-life methods (e.g., experience-sampling in addition to the EAR) to examine how much, how, and why people think and talk about their future in everyday life to create more ecologically valid measures of FTP.
With regard to the qualities of speech, we limited our exploration to a selection of six word categories based on the socioemotional selectivity theory (Carstensen et al., 1999) and our approach was completely exploratory. The newest version of LIWC contains over 80 word categories, including those related to biological processes, perceptual processes, and personal concerns. Future research might investigate these to shed more light on further functions of talking about the future.
Conclusions
Future time perspective has been widely studied in the literature in relation to healthy aging. However, past work was dominated by findings based on self-report. In this work, we explored whether FTP is associated with an objective and real-life verbal behavior, namely, talking about the personal future. Although this behavior was unrelated to general subjective views of the future, it was still associated with the life satisfaction of young and older adults. The life satisfaction of young adults was predicted by both their subjective perception of the future and by how much they talked about the future and used family-related words. For older adults, however, the most important predictor of life satisfaction was their subjective view of the future, followed by how much they used achievement-related words while talking about the future. This shows that for older adults, it is not how much they talk about their future in everyday life, but how they think about their future that matters. This suggests that self-report may be a more suitable method to assessing the future time perspective of older adults, whereas the future time perspective of young adults may be more diversely represented in different behaviors (e.g., thinking and talking) and suitable to be measured in alternative ways.
1Life satisfaction was assessed both at T1 and T2. A paired Mann-Whitney U-test showed that young adults had significantly higher life satisfaction at T2 (Median = 5.4) compared to T1 (Median = 5.2), Z = −2.31, p = .02. For older adults, no significant difference emerged (T1: Median = 5.2; T2: Median = 5.3), Z = −0.42, p = .68. Thus, for young adults, the relationships of FTP and talking behavior to T1 and T2 life satisfaction were examined separately. All predictor variables showed consistent relationships to life satisfaction at T1 and T2 (standardized betas for T1 given here; see Table 3 for T2): FTP (β = .29, p = .03), frequency of future-related utterances (β = .34, p = .05), positive emotion words (β = .01, p = .96), negative emotion words (β = −.04, p = .72), family words (β = .22, p = .09), friends words (β = .05, p = .69), affiliation words (β = .04, p = .61), achievement words (β = .09, p = .56). Given that the difference between T1 and T2 life satisfaction did not affect the patterns of relationships between variables, following the temporal logic of our conceptual model, we reported life satisfaction assessed at T2 (following FTP at T1 and real-life talking behavior between T1 and T2).
References
(2009). Time perspective, personality and smoking, body mass, and physical activity: An empirical study. British Journal of Health Psychology, 14, 83–105. https://doi.org/10.1348/135910708X299664
(2012). Forgivingness and subjective well-being in adulthood: The moderating role of future time perspective. Journal of Research in Personality, 46, 32–39. https://doi.org/10.1016/j.jrp.2011.11.004
(2002). Emotionally charged autobiographical memories across the life span: The recall of happy, sad, traumatic and involuntary memories. Psychology and Aging, 17, 636–652. https://doi.org/10.1037/0882-7974.17.4.636
(2004).
Balancing time perspective in pursuit of optimal functioning . In P. A. LinleyS. JosephEds., Positive psychology in practice (pp. 165–178). Hoboken, NJ: Wiley. https://doi.org/10.1002/9780470939338.ch10(2006). When does measurement invariance matter? Medical Care, 44, 176–181. https://doi.org/10.1097/01.mlr.0000245143.08679.cc
(2005). What is successful ageing and who should define it? BMJ, 331(7531), 1548–1551. https://doi.org/10.1136/bmj.331.7531.1548
(2017). Language-based personality: A new approach to personality in a digital world. Current Opinion in Behavioral Sciences, 18, 63–68. https://doi.org/10.1016/j.cobeha.2017.07.017
(2014). Measuring future time perspective across adulthood: Development and evaluation of a brief multidimensional questionnaire. The Gerontologist, 54, 1075–1088. https://doi.org/10.1093/geront/gnu076
(2016). Future time perspective and awareness of age-related change: Examining their role in predicting psychological well-being. Psychology and Aging, 31, 605–617. https://doi.org/10.1037/pag0000101
(1999). Taking time seriously. A theory of socioemotional selectivity. American Psychologist, 54, 165–181. https://doi.org/10.1037/0003-066X.54.3.165
(1996). Future orientation scale, Unpublished manuscript, Stanford University, Stanford, CA.
(2007). Testing models of the structure and development of future time perspective: Maintaining a focus on opportunities in middle age. Psychology and Aging, 22, 186–201. https://doi.org/10.1037/0882-7974.22.1.186
(2010). Life satisfaction, sense of coherence and job precariousness in Italian young adults. Journal of Adult Development, 17, 177–189. https://doi.org/10.1007/s10804-010-9099-2
(2011). Being well as time goes by: Future time perspective and well-being. International Journal of Psychology & Psychological Therapy, 11, 219–232.
(2011). Frequency, characteristics and functions of future-oriented thoughts in daily life. Applied Cognitive Psychology, 25, 96–103. https://doi.org/10.1002/acp.1647
(2014). Time since birth and time left to live: Opposing forces in constructing psychological wellbeing. Ageing & Society, 34, 1193–1218. https://doi.org/10.1017/S0144686X13000032
(2018). Conversational time travel: Evidence of a retrospective bias in real life conversations. Frontiers in Psychology, 9. https://doi.org/10.3389/fpsyg.2018.02160
(1985). The Satisfaction with Life Scale. Journal of Personality Assessment, 49, 71–75. https://doi.org/10.1207/s15327752jpa4901_13
(2014).
Language and interpersonal relationships . In T. M. HoltgravesEd., The Oxford handbook of language and social psychology (pp. 188–200). New York, NY: Oxford University Press.(2010). Using three levels of personality to predict time perspective. Current Psychology, 29, 95–103.
(1950). Childhood and society. New York, NY: Norton & Co.
(2017). Being present: Focusing on the present predicts improvements in life satisfaction but not happiness. Emotion, 17, 1047.
(1975). “Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–198. https://doi.org/10.1016/0022-3956(75)90026-6
(2016). The limits of a limited future time perspective in explaining age differences in emotional functioning. Psychology and Aging, 31, 583–593. https://doi.org/10.1037/pag0000060
(2015). Attachment relationships and life satisfaction during emerging adulthood. Social Indicators Research, 121, 833–847. https://doi.org/10.1007/s11205-014-0655-1
(2017). Smartphone sensing methods for studying behavior in everyday life. Current Opinion in Behavioral Sciences, 18, 83–90. https://doi.org/10.1016/j.cobeha.2017.07.018
(1972). Developmental tasks and education. New York, NY: McKay.
(2012). Positive affect, meaning in life, and future time perspective: An application of socioemotional selectivity theory. Psychology and Aging, 27, 181–189. https://doi.org/10.1037/a0023965
(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. https://doi.org/10.1080/10705519909540118
(2014).
Natural language use as a marker of personality . In T. M. HoltgravesEd., Oxford library of psychology. The Oxford handbook of language and social psychology (pp. 201–218). New York, NY: Oxford University Press.(2011).
Introduction . In R. B. KlineEd., Principles and practice of structural equation modeling (pp. 3–18). New York, NY: Guilford.(2018). Future time perspective: A systematic review and meta-analysis. Journal of Applied Psychology, 103, 867–893. https://doi.org/10.1037/apl0000306
(2008). Time perspective as a mechanism for functional developmental adaption. Journal of Social, Evolutionary, and Cultural Psychology, 2, 1–22. https://doi.org/10.1037/h0099336
(2010). Family as a salient source of meaning in young adulthood. The Journal of Positive Psychology, 5, 367–376. https://doi.org/10.1080/17439760.2010.516616
(2002). Time counts: Future time perspective, goals, and social relationships. Psychology and Aging, 17, 125–139. https://doi.org/10.1037/0882-7974.17.1.125
(2017).
Perceiving future time across adulthood . In G. OettingenA. SevincerP. GollwitzerEds., The psychology of thinking about the future (pp. 310–331). New York, NY: Guilford.(2000). Optimism, satisfaction and time perspective in the elderly. International Journal of Aging and Human Development, 51, 167–181.
(2018). Different future time perspectives interplay in predicting life satisfaction. GeroPsych, 31, 103–113. https://doi.org/10.1024/1662-9647/a000192
(2017). New evaluation of the electronically activated recorder (ear): Obtrusiveness, compliance, and participant self-selection effects. Frontiers in Psychology, 8, 658. https://doi.org/10.3389/fpsyg.2017.00658
(2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7, 64–82. https://doi.org/10.1037/1082-989X.7.1.64
(2014). iEAR (2.0). [iPhone application software]
(2001). The electronically activated recorder (EAR): A device for sampling naturalistic daily activities and conversations. Behavior Research Methods, Instruments, & Computers, 33, 517–523. https://doi.org/10.3758/BF03195410
(2012). Naturalistic observation of health-relevant social processes: The electronically activated recorder (EAR) methodology in psychosomatics. Psychosomatic Medicine, 74, 410–417. https://doi.org/10.1097/PSY.0b013e3182545470
(2018). “LIWC auf Deutsch”: The development, psychometrics, and introduction of DE-LIWC, https://doi.org/10.17605/OSF.IO/TFQZC
(2012). The smartphone psychology manifesto. Perspectives on Psychological Science, 7, 221–237. https://doi.org/10.1177/1745691612441215
(1976).
Age and the life course . In R. BinstockE. ShanasEds., Handbook of aging and social sciences (pp. 35–55). New York, NY: Van Nostrand Reinhold.(2015). Structural equation modeling reporting practices for language assessment. Language Assessment Quarterly, 12, 305–319. https://doi.org/10.1080/15434303.2015.1050101
(2015). Living in the past, present, and future: Measuring temporal orientation with language. Journal of Personality, 85, 270–280. https://doi.org/10.1111/jopy.12239
(2014).
Me and my stories . In T. HoltgravesEd., Handbook of language and social psychology (pp. 265–279). New York, NY: Oxford University Press.(1999). Linguistic Inquiry and Word Count (LIWC): A computer-based text analysis program. Mahwah, NJ: Erlbaum.
(2003). Words of wisdom: Language use over the life span. Journal of Personality and Social Psychology, 85, 291–301. https://doi.org/10.1037/0022-3514.85.2.291
(2014). Age differences in subjective well-being across adulthood: The roles of savoring and future time perspective. International Journal of Aging in Human Development, 78, 3–22. https://doi.org/10.2190/AG.78.1.b
(2011). The unpredictable past: Spontaneous autobiographical memories outnumber autobiographical memories retrieved strategically. Consciousness and Cognition, 20, 1842–1846. https://doi.org/10.1016/j.concog.2011.07.010
(1996). Affective ties to parents in early adulthood and self-esteem across 20 years. Social Psychology Quarterly, 59, 96–106.
(2017). A three-component model of future time perspective across adulthood. Psychology and Aging, 32, 597–607. https://doi.org/10.1037/pag0000191
(2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48, 1–36. https://doi.org/10.18637/jss.v048.i02
(1997). Successful aging. The Gerontologist, 37, 433–440. https://doi.org/10.1093/geront/37.4.433
. (2016). RStudio: Integrated development for R. Boston, MA: RStudio, Inc. Retrieved from http://www.rstudio.com/
(2003). Life scripts help to maintain autobiographical memories of highly positive, but not highly negative, events. Memory & Cognition, 31, 1–14. https://doi.org/10.3758/BF03196077
(2016). From past to future: Temporal self-continuity across the life span. Psychology and Aging, 31, 631–639. https://doi.org/10.1037/pag0000090
(2003).
SWLS – Satisfaction with Life Scale . In J. SchumacherA. KlaibergE. BrahlerEds., Diagnostische Verfahren zu Lebensqualitat und Wohlbefinden (pp. 305–309). Göttingen, Germany: Hogrefe.(2013). Personality, gender, and age in the language of social media: The open-vocabulary approach. PLoS One, 8, e73791. https://doi.org/10.1371/journal.pone.0073791
(2015). Extracting human temporal orientation from Facebook language. Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (pp. 409–419). https://doi.org/10.1111/jopy.12239
(1998). The importance of future time perspective in theories of motivation. The Journal of Psychology: Interdisciplinary and Applied, 132, 154–168. https://doi.org/10.1080/00223989809599156
(1992). The apprehensive respondent: Failing to rate future life satisfaction in older adults. Psychology and Aging, 7, 484–486.
(2004). Placing motivation and future time perspective theory in a temporal perspective. Educational Psychology Review, 16, 121–139. https://doi.org/10.1023/B:EDPR.0000026609.94841.2f
(2006). The search for “optimal” cutoff properties: Fit index criteria in structural equation modeling. The Journal of Experimental Education, 74, 267–288. https://doi.org/10.3200/JEXE.74.3.267-288
(2016). Hourglass half-full or half-empty? Future time perspective and preoccupation with negative events across the life span. Psychology and Aging, 31, 558–573. https://doi.org/10.1037/pag0000097
(2007). Selective self-presentation in computer-mediated communication: Hyperpersonal dimensions of technology, language, and cognition. Computers in Human Behavior, 23, 2538–2557. https://doi.org/10.1016/j.chb.2006.05.002
(2006). Parents influence on children’s future orientation and saving. Journal of Economic Psychology, 27, 140–164. https://doi.org/10.1016/j.joep.2005.06.016
(1999). Putting time in perspective: A valid, reliable individual-difference metric. Journal of Personality and Social Psychology, 77, 1271–1288. https://doi.org/10.1037/0022-3514.77.6.1271
Appendix
