The Illusion of Time: Testing the Bidirectional Relationship Between Belief in Free Will and Temporal Horizons

Abstract

The current research examined the bidirectional relationship between belief in free will and perceptions of time. Studies 1 and 2 explored the association between free will beliefs (FWB) and future time perspective (FTP) in correlational and longitudinal studies, respectively. Results indicated a positive relationship between the two constructs. Study 3 manipulated FTP and assessed FWB. Study 4 manipulated FWB and assessed FTP about current goal-pursuit. While manipulating FTP did not directly influence FWB, attenuating FWB led participants to perceive time as more limited to accomplish their goals. Exploratory mediation suggests that the perception of increased paths to successful goal-pursuit accounts for this relationship. The interplay between FWB and FTP is discussed.

Keywords

belief in free will future time perspective perceived opportunities temporal horizons

Philosophers, psychologists, and even laypersons have debated about the existence of free will. On one end of the spectrum is the notion that free will is an illusion (Wegner, 2004) or the mistaken assumption of conscious causation (Bargh, 2008). On the other end is a conviction that individuals have the ability to freely behave in accordance to their own values (Nahmias et al., 2005; Sarkissian et al., 2010). Regardless of whether free will exists, the belief in free will may influence our thoughts and actions (Monroe & Malle, 2010). In fact, many lines of research attest to the importance of believing in free will (e.g., Baumeister & Brewer, 2012). In the present research, we investigated how belief in free will related to perceptions of time. Specifically, we tested whether there is a bidirectional relationship between free will beliefs (FWB) and future time perspective (FTP).

FWB

While scholars operationalize free will in different terms, most people understand free will as the ability to choose one’s actions and determine one’s outcomes (Feldman et al., 2014; Stillman et al., 2011). Seminal research has found that attenuating FWB increases cheating (Vohs & Schooler, 2008), aggression (Baumeister et al., 2009), and conformity (Alquist et al., 2013). Moreover, believing in free will influences our causal attributions when judging others’ behavior (Genschow et al., 2017, 2019) and increases the severity of punishment toward people who commit immoral acts (Clark et al., 2014; Shariff et al., 2014).

Inherent in the belief in free will is the notion that people are active agents. That is, there is a sense of authorship behind one’s choices. Research has found that disbelief in free will attenuates components of self-agency (Lynn et al., 2014), and free will increases when people perceive they are producing chosen outcomes while engaging in goal-directed behaviors (Aarts & van den Bos, 2011). Furthermore, belief in free will is associated with proximal determinants of agency such as higher self-efficacy, greater mindfulness, and the pursuit of meaningful goals (Crescioni et al., 2016). These findings suggest that FWB invoke personal agency and heighten the experience of action control.

FTP

FTP represents people’s experience of time left to complete one’s goals (Lang & Carstensen, 2002). Time can feel limited or constrained, as if we are running out of time, or expansive or open-ended, as if there is ample time to pursue opportunities in front of us. These subjective time horizons are largely motivational processes and a key component in socioemotional selectivity theory (SST; Carstensen, 2006). SST proposes that people have a unique ability to monitor time, and the types of goals they pursue are influenced by FTP (Carstensen et al., 1999; Fung & Carstensen, 2004). When time is perceived as limited, people prioritize more emotionally meaningful goals. When time is perceived as expansive, people focus on preparatory goals. In fact, SST suggests that people’s subjective perceptions of time as limited or expansive are responsible for changes in motivation perhaps above and beyond age itself (Fung et al., 1999; Löckenhoff & Carstensen, 2004).

In addition to the prioritization of goals (Lang & Carstensen, 2002), FTP influences emotional functioning (Grühn et al., 2016), subjective well-being (Allemand et al., 2012; Hicks et al., 2012), and decision-making (Rudd et al., 2012), with the general pattern suggesting that an expansive time perspective enables more positive psychological consequences. More recently, Kooij and colleagues (2018) conducted a meta-analysis and found that open-ended time horizons were positively associated with achievement, psychological well-being, and health-improving behaviors. Overall, these findings suggest that shifting people’s temporal horizons influences a host of thoughts and behaviors.

The Bidirectionality of FWB and FTP

FWB and FTP have similar agentic and motivational qualities influencing our thoughts and actions. FWB serve as an agentic tool that is associated with traits such as self-efficacy, locus of control, and self-esteem (Feldman, 2017), motivating one to pursue personally meaningful goals and outcomes (Crescioni et al., 2016; Dennett, 2003). Similarly, expansive FTP can enhance a sense of agency by inducing thoughtful reflections of potential courses of actions (Carstensen, 2006). In fact, research has shown that open-ended FTP is positively related to the same agentic traits of self-efficacy, locus of control, and self-esteem as FWB (Kooij et al., 2018), and people who set plans for the future also tend to believe “they” have the ability to influence future goals (Shipp et al., 2009).

Given the similar agentic and motivational processes attached to both FWB and FTP, we propose that these constructs are strongly linked to each other. Specifically, we believe our sense of agency will wax and wane as a function of our perceived temporal horizons and vice versa. To the extent that belief in free will is associated with enhanced perception that one can control and overcome one’s current situation (e.g., Aarts & van den Bos, 2011), it should expand perceived time horizons. Further, perceived time horizons should directly bear on our sense of FWB. Specifically, when time is perceived as expansive and slowing down, people are more likely to feel they have the opportunity to enact their own actions among infinite possibilities (e.g., Lang & Carstensen, 2002) and, as a result, experience an enhanced sense of free will in pursuing their goals.

As previously suggested, one plausible mechanism for this hypothesized bidirectional relationship is that both FWB and FTP are related to the perception that one can successfully attain important goals (Crescioni et al., 2016; Shipp et al., 2009). People who believe in free will hold the perception that they can control their current situation, potentially leading to more perceived pathways to goal attainment (e.g., Aarts & van den Bos, 2011). Similarly, when time is perceived as expansive, people perceive more plausible routes to goal attainment (e.g., even if one route is unsuccessful, there is still time to pursue different strategies; Davis & Hicks, 2013). Thus, both FWB and FTP are connected to the belief that one has the opportunity to enact different strategies to pursue goals (e.g., Lang & Carstensen, 2002). We believe this unique connection between one’s sense of agency and perception of enhanced opportunities to accomplish one’s goals underlies the relationship between FWB and FTP.

Overview of the Current Studies

In the present research, we examined the association between FWB and FTP across four studies.¹ Study 1 was a correlational study that tested whether FWB are positively associated with expansive FTP. Study 2 was a longitudinal study that examined the reciprocal relationship between FWB and FTP over time. Studies 3 and 4 were experiments assessing how manipulating people’s FTP (Study 3) or FWB (Study 4), influenced their FWB or FTP, respectively. Studies 3 and 4 also explored two potential mediators, sense of agency and perceived opportunities, that might explain the predicted bidirectional relationship between FWB and FTP.

Sample Size Determination

In Studies 1 and 2, the sample size determination was guided by the recommendation of 250 minimum participants for obtaining stable correlation coefficients (Schönbrodta & Perugini, 2013). For Study 2, we also performed a priori power analysis based on 1,000 Monte Carlo replication simulations using Webpower (Zhang & Yuan, 2018), which revealed a minimum sample size of 265 to detect a medium size of β estimates (.15) with a power equal to .95 at an α level of .05. In expecting attrition during the three waves of surveys, we set an overall goal to maximize power by recruiting more than 300 participants and ended up recruiting the final sample size using available lab resources. In Study 3, we conducted a priori power analysis to achieve a power equal to .95 with a medium effect size (d = .60) at an α level of .05 based on the FTP manipulation by Davis and Hicks (2013). A sample size of 122 was required. In Study 4, we conducted a priori power analysis to achieve a power equal to .95 with a medium effect size (d = .50) at an α level of .05. A sample size of 176 was required. Data collection was terminated once these goals were met.

Study 1

Study 1 tested our prediction that FWB are positively correlated with open-ended FTP.

Method

Participants

Two hundred and ninety-two participants (212 females, 76 males, 1 transgender, and 3 unreported; M _age = 20.06, SD = 2.09) recruited from a large university completed the study for course credit. Participants were predominantly White (76%) and non-Hispanic (75%).

Materials and Procedure

Participants completed the measures described in order below and were debriefed following completion of the study.

FWB

To assess FWB, participants completed the 7-item Free Will subscale of the Free Will and Determinism Plus (FAD+) Scale (Paulhus & Carey, 2011). Responses (e.g., “People have complete control over the decisions they make.”) were made on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Items were averaged to produce a composite score, where higher values indicated greater FWB.

FTP

To assess FTP, participants completed the 10-item FTP Scale (Carstensen & Lang, 1996). Responses (e.g., “I have a sense that time is running out” [reverse coded]) were made on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Items were averaged to produce a composite score, where higher values indicated more open-ended FTP.

Results

Descriptive statistics and reliabilities for all studies can be found in Table 1. Supporting our prediction, bivariate correlations revealed a significant positive relationship between FWB and FTP, r(290) = .35, p < .001.^2,3 The greater belief in free will, the more participants perceived time to be open-ended and expansive.

Table 1.

Descriptive Statistics and Reliabilities of All Variables Across Studies.

Variable	Mean	SD	Cronbach’s α
Study 1 (n = 392)
FWB	5.02	0.90	.80
FTP	5.00	0.81	.81
Study 2: Wave 1 (n = 378)
FWB	5.36	0.92	.77
FTP	5.23	0.82	.80
Study 2: Wave 2 (n = 335)
FWB	5.30	0.92	.83
FTP	5.13	0.80	.82
Study 2: Wave 3 (n = 300)
FWB	5.23	0.94	.85
FTP	5.16	0.85	.84
Study 3 (n = 124)
FTP manipulation check	3.78	1.40	.82
FWB	4.37	1.07	.82
Sense of agency	5.75	1.20	.73
Pathway opportunities	6.01	1.06	.72
Study 4 (n = 198)
FWB manipulation check	5.14	1.58	.85
FTP of goals	4.19	1.17	.88
Sense of agency	5.38	1.06	.90
Pathway opportunities	5.35	1.17	.92

Note. Response rates for each wave of Study 2 were 100% (378 of 378) at Wave 1, 88.6% (335 of 378) at Wave 2, and 79.4% (300 of 378) at Wave 3. FWB = free will beliefs; FTP = future time perspective.

Study 2

Study 2 examined the bidirectional relationship between FWB and FTP over time in a longitudinal study. We predicted that FWB is a precursor to expansive FTP and vice versa.

Method

Participants

Three hundred and seventy-eight participants (290 females, 86 males, and 2 unreported; M _age = 18.58, SD = 1.06) from a large university completed a three-wave longitudinal study for course credit. Participants were predominantly White (79%) and non-Hispanic (75%).

Materials and Procedure

Participants received and completed each of the three surveys online, with 3-week time lags across waves. We employed this design with the 3-week time lag in order to provide broad coverage of a single semester while keeping the overall study short enough to minimize participant attrition. The surveys in each wave included FWB and FTP in order and other measures irrelevant to the present study.

FWB and FTP

FWB and FTP were measured using the same scales from Study 1. Composite scores were computed for FWB and FTP at each wave.

Attrition

As presented in Table 1, there was participant attrition; however, attrition analyses revealed that there were no significant differences in FWB (M _{no drop} = 5.36 vs. M _drop = 5.35), t(376) = 0.129, p = .897, d = .01, and FTP (M _{no drop} = 5.23 vs. M _drop = 5.21), t(376) = 0.211, p = .833, d = .02, reported at Wave 1 between participants who completed all the surveys (n = 284) and those who dropped out from any wave (n = 94).

Results

To examine the relationship between FWB and FTP over time, we conducted the longitudinal analyses by using a random intercepts cross-lagged panel model (RI-CLPM; Hamaker et al., 2015). As an alternative to traditional CLPM, the RI-CLPM allows for testing the within-person relationships between constructs. RI-CLPM specifically helps minimize a possibility that stable “trait-like” between-person variance contaminates within-person cross-lagged effects, producing less biased cross-lagged estimates (Hamaker et al., 2015). We used Mplus (Version 8.0; Muthén & Muthén, 1998–2012) to estimate the RI-CLPM model (see Supplementary Materials for analytic approach).

Estimated correlation coefficients between FWB and FTP across waves are presented in Table 2. Consistent with Study 1, FWB were positively correlated with FTP across waves. We first conducted the unconstrained RI-CLPM and found that it fits the data well, χ²(5) = 15.28, p = .009; root mean square error of approximation (RMSEA) = .074, 90% confidence interval (CI) [0.033, 0.117]; comparative fit index (CFI) = .989. In this model, the separate autoregressive and cross-lagged paths were freely estimated. For example, the autoregressive path from FWB at Wave 1 to FWB at Wave 2 was freely estimated as the path from FWB at Wave 2 to FWB at Wave 3, and the cross-lagged path from FWB at Wave 1 to FTP at Wave 2 was freely estimated as the path from FWB at Wave 2 to FTP at Wave 3. Given the assumption there would be stability of the autoregressive and cross-lagged associations between FWB and FTP across adjacent waves, we conducted the constrained RI-CLPM by imposing equality constraints on the autoregressive and cross-lagged paths. The constrained model fit the data well, χ²(9) = 21.06, p = .012; RMSEA = .06, 90% CI [0.026, 0.093]; CFI = .987, and the additional constraints did not worsen fit and therefore produced a more parsimonious model, Δχ²(4) = 5.78, critical value = 9.49. This result justifies our assumption about the stability of the autoregressive and cross-lagged paths across adjacent intervals. In this constrained model, the cross-lags from FWB to FTP and from FTP to FWB were significant (ps < .008), providing support for the reciprocal association between FWB and FTP.

Table 2.

Estimated Correlation Coefficients Between FWB and FTP Across Waves in Study 2.

Measure	1	2	3	4	5	6
1. FWB₁	—
2. FWB₂	.67***	—
3. FWB₃	.61***	.72***	—
4. FTP₁	.25***	.23***	.16**	—
5. FTP₂	.27***	.31***	.29***	.69***	—
6. FTP₃	.20***	.23***	.33***	.61***	.73***	—

Note. The subscript numbers attached to FWB and FTP indicate wave numbers. FWB = free will beliefs; FTP = future time perspective.

**p < .01. ***p < .001.

As shown in Figure 1, FWB at a previous wave were a statistically significant precursor of FTB at a later wave (e.g., β = .285 predicting FTP at Wave 2 from FWB at Wave 1) and vice versa (e.g., β = .32 predicting FWB at Wave 2 from FTP at Wave 1). As often found in the RI-CLPM, the autoregressive paths in FWB and FTP were not statistically significant (ps > .052), which are caused by the fact that the RI-CLPM separated the time-invariant between-person stability from the temporal within-person stability (Hamaker et al., 2015).

Figure 1.

Constrained random intercept (RI) cross-lagged panel model (Hamaker et al., 2015) of the reciprocal relationship between free will beliefs (FWB) and future time perspective (FTP) across three waves with 3-week time lags in Study 2. Squares represent observed variables, and circles denote latent variables (RI represents random intercepts). Values attached to arrows between the same latent variables (e.g., FWB₁ to FWB₂) indicate standardized autoregressive parameter estimates; values attached to arrows between different latent variables (e.g., FWB₁ to FTP₂) indicate standardized cross-lagged parameter estimates; values given in brackets indicate 95% confidence interval. The subscript numbers attached to FWB and FTP indicate wave numbers. ^† p < .10. *p < .01. **p < .001.

Study 3

Study 2 established a bidirectional relationship between FWB and FTP over time. In Study 3, we examined one direction of the causal relationship by manipulating FTP and then assessing FWB. We predicted that manipulating FTP would lead to a corresponding change in people’s belief that they possess free will. Additionally, we explored potential mechanisms for the bidirectional relationship between FWB and FTP by examining one’s sense of agency and pathway opportunities.

Method

Participants

One hundred and twenty-four participants (84 females, 40 males; M _age = 18.85, SD = 1.09) recruited from a small college completed the study for course credit. Participants were predominantly White (66%) and non-Hispanic (94%).

Materials and Procedure

Participants completed the measures described below and were debriefed following completion of the study.

FTP manipulation

Participants completed an FTP manipulation (Davis & Hicks, 2013, Study 2). They were randomly assigned to read a passage describing how people have more or less time than they think to accomplish important goals for the open-ended and limited FTP conditions, respectively. Participants were then asked to describe three important goals and corresponding reasons why they have plenty (don’t have enough) time to accomplish them.

Manipulation check

To ensure that open and limited FTP was induced successfully, participants indicated their agreement with two face-valid statements (e.g., “Overall, I feel I have a limited amount of time to accomplish my goals and pursue opportunities” [reverse coded]). Both items used a 7-point scale (1 = strongly disagree, 7 = strongly agree) and were averaged to produce a composite score.

FWB

FWB were measured using the same scale from Studies 1 and 2. Responses were averaged to produce a composite score.

Hope

Participants completed a 6-item State Hope Scale (Snyder et al., 1996). Three items assessed the theorized agency component of hope (e.g., “At the present time, I am energetically pursuing my goals.”), which reflected the sense of agency mediator. Three items assessed the theorized pathways component of hope (e.g., “I can think of many ways to reach my goals.”), which reflected the pathway opportunities mediator. Responses were made on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Items were averaged to produce composite agency and pathway opportunities scores.

Results

Main Analyses

We first conducted an independent samples t test to assess the FTP manipulation. There was a significant difference in FTP, t(122) = 4.78, p < .001, d = .85, 95% CI [0.65, 1.56]. Participants in the open-ended time condition reported a more open, expansive time horizon (M = 4.33, SD = 1.14) than those in the limited time condition (M = 3.23, SD = 1.42), suggesting the manipulation was successful.

For our main analyses, we conducted an independent samples t test to determine whether there was a difference in FWB. Unexpectedly, there was no significant difference in FWB, t(122) = 0.20, p = .839, d = .04, 95% CI [−0.34, 0.42], with participants in the open-ended time condition reporting as much FWB (M = 4.39, SD = 0.98) as those in the limited time condition (M = 4.35, SD = 1.15).⁴

Exploratory Mediation Analyses

Guided by recent discussions on mediation analysis that a significant total effect is not a necessary condition for observing a significant indirect effect when a mediator is theoretically meaningful (e.g., Rucker et al., 2011), we tested whether FTP mediated the effect of the manipulation on FWB using the Hayes’s PROCESS macro (Model 4; Hayes, 2018). As shown in Figure 2, the mediation analysis with 10,000 bootstrap sampling (Preacher & Hayes, 2008) revealed a significant indirect effect of the FTP manipulation on FWB through FTP. Supporting our prediction, this finding suggests that although our manipulation did not directly influence FWB, participants who perceived their time as more expansive (vs. limited) reported that they possess free will to a greater (vs. lesser) extent.

Figure 2.

Simple mediation analysis: the indirect effect of the future time perspective manipulation on belief in free will through expansive future time perspective in Study 3. The expansive future time perspective condition was coded as 1, and the limited future time perspective condition was coded as 0. The value in the parentheses indicates the total effect of the manipulation on belief in free will.

Mediating Roles of a Sense of Agency and Pathway Opportunities

To explore the possibility that FTP is linked with FWB through a sense of agency and enhanced perceived opportunities to attain future goals, we tested additional mediation analyses. Specifically, we first conducted a multiple mediation analysis in which both a sense of agency and pathway opportunities were entered as mediators simultaneously to explain the path from FTP to FWB. We again tested the indirect effects by using the percentile bootstrap estimation approach with 10,000 samples through the PROCESS macro (Hayes, 2018). As illustrated in Figure 3, the results showed that there was a significant indirect effect through pathway opportunities, but not through agency. The nonsignificant indirect effect via agency was mainly because FTP was not correlated with the agency, r = .10, p = .265.

Figure 3.

Multiple mediation analysis: the indirect effects of expansive future time perspective on belief in free will through a sense of agency and pathway opportunities in Study 3. The value in the parentheses indicates the total effect of expansive future time perspective on belief in free will.

To further demonstrate the mediating role of the pathway opportunities in the effect of FTP on FWB, we conducted a serial mediation analysis by using 10,000 bootstrap samples through the PROCESS macro (Model 6; Hayes, 2018). Specifically, the serial mediation tested whether the pathway opportunities mediated the effect of FTP induced by the manipulation on FWB (see Figure 4). As expected, we found a significant indirect effect only through the serial mediation of the FTP and pathway opportunities, as its bootstrapping 95% CI did not include zero (see Table 3).⁵ Consistent with our prediction, these findings suggest that FTP is connected with FWB via the perceptions of increased opportunities to attain goals.

Figure 4.

Serial mediation analysis: the indirect effect of the future time perspective manipulation on belief in free will through expansive future time perspective and pathway opportunities in Study 3. The expansive future time perspective condition was coded as 1, and the limited future time perspective condition was coded as 0. The value in the parentheses indicates the total effect of the manipulation on belief in free will.

Table 3.

Coefficients and Bootstrapping 95% Confidence Interval for the Indirect Effects of the Serial Mediation in Study 3.

Effects	Product of Coefficient		Bootstrapping 95% Confidence Interval
Effects	Estimate	SE	[Lower, Upper]
Total indirect effects	.168	.107	[−.038, .382]
Condition → FTP → FWB	.160	.093	[−.009, .354]
Condition → pathway opportunities → FWB	−.056	.067	[−.206, .059]
Condition → FTP → pathway opportunities → FWB	.064	.036	[.007, .146]

Note. FWB = free will beliefs; FTP = future time perspective.

Study 4

Studies 1–3 established a consistent relationship between FWB and FTP. While Study 3 failed to find evidence that manipulating FTP directly influenced FWB, we did find that the manipulation led to an indirect effect on FWB through FTP. Exploratory analysis also revealed that perceptions of increased opportunities to accomplish goals mediated the relationship between FTP and FWB. In Study 4, we sought to examine whether manipulating FWB would influence FTP of goals. We again tested for the mediating effect of sense of agency and pathway opportunities to instill more confidence in our findings.

Method

Participants

One hundred and ninety-eight participants (110 females, 88 males), recruited from Amazon’s Mechanical Turk, completed the study and were compensated with US$1.50.⁶ Participants were from the United States, diverse in age (M _age = 36.81, SD = 12.60), and predominantly White (72%) and non-Hispanic (89%).

Materials and Procedure

Participants completed the measures described below and were debriefed following completion of the study.

FWB manipulation

Participants completed a free will manipulation (Seto & Hicks, 2016). They were randomly assigned to read a description of high or low FWB and were presented with statements that reflect FWB. Participants were asked to think about why these statements are true or false based on their own experiences and select three statements that have proven especially true or false in their life for high and low FWB conditions, respectively. They were then asked to describe how the statements were true or false based on their experiences.

Manipulation check

To ensure that high and low FWB were induced successfully, participants indicated their agreement with two face-valid statements. One item reflected general FWB (“People have complete free will”; FAD+; Paulhus & Carey, 2011), and the other item reflected personal FWB (“I have complete free will”; FAD Scale; Rakos et al., 2008). Both items used a 7-point scale (1 = strongly disagree, 7 = strongly agree) and were averaged to produce a composite score, where higher values indicated greater FWB.

Goal description task

Participants were asked to think about “some concrete goals [they] are trying to accomplish.” Then, they were asked to describe three different goals and answer questions related to each goal.

FTP of goals

To assess FTP in relation to goal-pursuit, participants completed a 5-item adapted version of the FTP Scale (Lang & Carstensen, 2002). Responses (e.g., “I have many opportunities to accomplish this goal in the future.”) were made on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Items were averaged to produce a composite score, where higher values indicated more open-ended FTP for goals.

Hope

Participants completed a 6-item adapted version of the State Hope Scale (Snyder et al., 1996) from Study 3. Items were averaged to produce composite agency and pathway opportunities scores.

Results

Main Analyses

We conducted an independent samples t test to assess our free will manipulation. There was a significant difference in FWB, t(158) = −10.12, p < .001, d = −1.44, 95% CI [−2.23, −1.50]. Participants in the high FWB condition reported greater FWB (M = 6.04, SD = 0.97) than those in the low FWB condition (M = 4.18, SD = 1.55), suggesting the manipulation was successful.

For our main analyses, we conducted an independent samples t test to determine whether there was a difference in FTP of goals. As shown in Figure 5, results indicated a significant difference in FTP, t(196) = −2.38, p = .018, d = −.34, 95% CI [−0.72, −0.07], such that participants who wrote about experiences involving low FWB reported less time to accomplish their goals (M = 3.99, SD = 1.27) than those who wrote about experiences with high FWB (M = 4.38, SD = 1.04).

Figure 5.

Differences in future time perspective of goals by condition in Study 4. Standard errors are represented by the error bars.

Mediating Roles of a Sense of Agency and Perceived Opportunities

We again explored the mediating role of a sense of agency and pathway opportunities to attain goals. We conducted a multilevel multiple mediation, given that hope evaluation and FTP for each goal were nested within individuals, by using the MLmed macro for SPSS (Rockwood & Hayes, 2017). Specifically, because our data have two levels (i.e., Level 1 for hope evaluation and FTP, Level 2 for condition), we estimated a 2-(1,1)-1 multilevel multiple mediation model. As outlined by Bauer et al. (2006), both the within- and between-person effects were estimated, including random effects, and the indirect effect was estimated through a Monte Carlo simulation generating 95% CI based on 10,000 resamples. As illustrated in Figure 6, there were significant within- and between-person effects of agency and pathway opportunities on FTP (except for the negative within-person effect of agency on FTP).⁷ Importantly, we found a significant between-person indirect effect of the FWB manipulation on FTP only through pathway opportunities (no within-person indirect effect specified as the independent variable was a between-subject factor), but not through a sense of agency, which is consistent with the findings of Study 3. In sum, these findings suggest that the difference in the FTP of goals by condition was mediated by the perceptions of enhanced opportunities to attain the goals rather than agentic feelings about achieving the goals.

Figure 6.

Multilevel multiple mediation analysis: the indirect effects of the belief in free will manipulation on expansive future time perspective through a sense of agency and pathway opportunities of goals in Study 4. The high belief in free will condition was coded as 1, and the low belief in free will condition was coded as 0. The value in the parentheses indicates the total effect of the manipulation on expansive future time perspective.

Discussion

The current research examined the relationship between FWB and FTP. Study 1 found a positive association between the two constructs such that greater FWB were associated with a more open-ended, expansive time horizon. Study 2 extended this finding in a longitudinal study. In Study 3, we found that manipulating FTP had an indirect influence on FWB. In Study 4, we found that manipulating FWB directly influenced FTP. Specifically, attenuating FWB led to a more limited, constrained time horizon. Our last two studies also identified a potential mediator for this relationship, the enhanced perceptions of opportunities. These studies suggest a bidirectional relationship between our FWB and subjective experience of time.

The agentic and motivational characteristics of FWB and FTP naturally link these two constructs. FWB are analogous to making choices (Feldman et al., 2014) and exercising control over one’s actions (Stillman et al., 2011). When people have this augmented view of agency, they experience time as open-ended, and opportunities to accomplish goals amplify in conjunction. Moreover, the conceptualization of FTP as a mechanism by which people anticipate and structure one’s future (Carstensen, 2006) was associated with FWB. When time horizons feel open, there is an enhanced perception that individuals can enact numerous actions in any situation, presumably because opportunities appear more infinite. Our exploratory mediation in Studies 3 and 4 supports these propositions. In both studies, we found that the salience of pathway opportunities mediated both directional relationships. Enhanced FWB highlight the boundless routes to goal attainment, making time appear expansive. Similarly, open-ended time enhances perceived opportunities to achieve goals, heightening the experience of action control.

Although a robust correlational relationship emerged between the two constructs, we only found experimental evidence that FWB directly influenced FTP. Given the motivational processes inherent in FTP and its influence on well-being and behavior (Kooij et al., 2018), it was surprising that shifting temporal horizons did not directly alter changes in FWB (although the manipulation indirectly influenced FWB via FTP). Perhaps, reflecting on goals in the abstract future is not strong enough to induce changes in FWB, as a myriad of research has shown that most people believe in free will (Nahmias et al., 2005). Another explanation may lie with the age of the sample. Cate and John (2007) found that a focus on opportunities was higher in young adulthood, while a focus on limitations was more prominent in late middle age. As Study 3 consisted solely of participants in young adulthood, it is possible that the more nuanced focus on opportunities inhibited fluctuations in FWB. Future research should explore this causal direction further with changes in the FTP manipulation or different age groups.

While our findings support the bidirectionality of FWB and FTP, future research is needed to address the limitations of the current research. For example, indirect effects through pathway opportunities between FWB and FTP found in Studies 3 and 4 should be interpreted with caution, as the temporal precedence among these variables cannot be concluded from the cross-sectional analysis (Maxwell & Cole, 2007). Future research should examine the mediation process by employing longitudinal studies and experimental-causal-chain designs (e.g., Spencer et al., 2005). Moreover, our samples were derived from the United States and likely represent WEIRD participants (Western, educated, industrialized, rich, and democratic societies; Henrich et al., 2010). Future research should examine different cultural groups to test the generalizability of our findings.

Conclusion

While empirical support for the existence of free will is scarce, the belief in free will is widespread and has vast implications for our thoughts, feelings, and behaviors. The present research uncovered a dynamic interplay between FWB and FTP such that agentic choice is linked to our subjective time horizons. Importantly, we demonstrate that the illusion of time can boost FWB, and the degree of FWB in our everyday lives is powerful enough to change the most arguably uncontrollable outcome, time.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Elizabeth Seto

Supplemental Material

The supplemental material is available in the online version of the article.

Notes

References

Aarts

van den Bos

(2011). On the foundations of beliefs in free will: Intentional binding and unconscious priming in self-agency. Psychological Science, 22, 532–537.

Allemand

Hill

P. L.

Ghaemmaghami

Martin

(2012). Forgivingness and subjective well-being in adulthood: The moderating role of future time perspective. Journal of Research in Personality, 46, 32–39.

Alquist

J. L.

Ainsworth

S. E.

Baumeister

R. F.

(2013). Determined to conform: Disbelief in free will increases conformity. Journal of Experimental Social Psychology, 49, 80–86.

Bargh

J. A.

(2008). Free will is un-natural. In Baer

Kaufman

J. C.

Baumeister

R. F.

Baer

Kaufman

J. C.

Baumeister

R. F.

(Eds.), Are we free? Psychology and free will (pp. 128–154). Oxford University Press.

Bauer

D. J.

Preacher

K. J.

Gil

K. M.

(2006). Conceptualizing and testing random indirect effects and moderated mediation in multilevel models: New procedures and recommendations. Psychological Methods, 11, 142–163.

Baumeister

R. F.

Brewer

L. E.

(2012). Believing versus disbelieving in free will: Correlates and consequences. Social and Personality Psychology Compass, 6, 736–745.

Baumeister

R. F.

Masicampo

E. J.

DeWall

C. N.

(2009). Prosocial benefits of feeling free: Disbelief in free will increases aggression and reduces helpfulness. Personality and Social Psychology Bulletin, 35, 260–268.

Carstensen

L. L.

(2006). The influence of a sense of time on human development. Science, 312, 1913–1915.

Carstensen

L. L.

Isaacowitz

D. M.

Charles

S. T.

(1999). Taking time seriously: A theory of socioemotional selectivity. American Psychologist, 54, 165–181.

10.

Carstensen

L. L.

Lang

F. R.

(1996). Future time perspective scale. Unpublished manuscript, Stanford University.

11.

Cate

R. A.

John

O. P.

(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.

12.

Clark

C. J.

Luguri

J. B.

Ditto

P. H.

Knobe

Shariff

A. F.

Baumeister

R. F.

(2014). Free to punish: A motivated account of free will belief. Journal of Personality and Social Psychology, 106, 501–513.

13.

Crescioni

A. W.

Baumeister

R. F.

Ainsworth

S. E.

Ent

Lambert

N. M.

(2016). Subjective correlates and consequences of belief in free will. Philosophical Psychology, 29, 1–23.

14.

Davis

W. E.

Hicks

J. A.

(2013). Maintaining hope at the 11th hour: Authenticity buffers the effect of limited time perspective on hope. Personality and Social Psychology Bulletin, 39, 1634–1646.

15.

Dennett

D. C.

(2003). The self as a responding—and responsible—artifact. Annals of the New York Academy of Sciences, 1001, 39–50.

16.

Feldman

(2017). Making sense of agency: Belief in free will as a unique and important construct. Social and Personality Psychology Compass, 11, 1–15.

17.

Feldman

Baumeister

R. F.

Wong

K. F. E.

(2014). Free will is about choosing: The link between choice and the belief in free will. Journal of Experimental Social Psychology, 55, 239–245.

18.

Fung

H. H.

Carstensen

L. L.

(2004). Motivational changes in response to blocked goals and foreshortened time: Testing alternatives to socioemotional selectivity theory. Psychology and Aging, 19, 68–78.

19.

Fung

H. H.

Carstensen

L. L.

Lutz

A. M.

(1999). Influence of time on social preferences: Implications for life-span development. Psychology and Aging, 14, 595–604.

20.

Genschow

Rigoni

Brass

(2017). Belief in free will affects causal attributions when judging others’ behavior. Proceedings of the National Academy of Sciences, 114, 10071–10076.

21.

Genschow

Rigoni

Brass

(2019). The hand of god or the hand of Maradona? Believing in free will increases perceived intentionality of others’ behavior. Consciousness and Cognition, 70, 80–87.

22.

Grühn

Sharifian

Chu

(2016). The limits of a limited future time perspective in explaining age differences in emotional functioning. Psychology and Aging, 31, 583–593.

23.

Hamaker

E. L.

Kuiper

R. M.

Grasman

R. P.

(2015). A critique of the cross-lagged panel model. Psychological Methods, 20, 102–116.

24.

Hayes

. (2018). Introduction to mediation, moderation, and conditional process analysis. Guilford.

25.

Henrich

Heine

S. J.

Norenzayan

(2010). Most people are not WEIRD. Nature, 466, 29.

26.

Hicks

J. A.

Trent

Davis

W. E.

King

L. A.

(2012). Positive affect, meaning in life, and future time perspective: An application of socioemotional selectivity theory. Psychology and Aging, 27, 181–189.

27.

Kooij

D. T. A. M.

Kanfer

Betts

Rudolph

C. W.

(2018). Future time perspective: A systematic review and meta-analysis. Journal of Applied Psychology, 103, 867–893.

28.

Lang

F. R.

Carstensen

L. L.

(2002). Time counts: Future time perspective, goals, and social relationships. Psychology and Aging, 17, 125–139.

29.

Löckenhoff

C. E.

Carstensen

L. L.

(2004). Socioemotional selectivity theory, aging, and health: The increasingly delicate balance between regulating emotions and making tough choices. Journal of Personality, 72, 1395–1424.

30.

Lynn

M. T.

Muhle-Karbe

P. S.

Aarts

Brass

(2014). Priming determinist beliefs diminishes implicit (but not explicit) components of self-agency. Frontiers in Psychology, 5, 1–8.

31.

Maxwell

S. E.

Cole

D. A.

(2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12, 23–44.

32.

Monroe

A. E.

Malle

B. F.

(2010). From uncaused will to conscious choice: The need to study, not speculate about people’s folk concept of free will. European Review of Philosophy, 1, 211–224.

33.

Muthén

L. K.

Muthén

B. O

. (1998–2012). Mplus user’s guide (8th ed.). Author.

34.

Nahmias

Morris

Nadelhoffer

Turner

(2005). Surveying freedom: Folk intuitions about free will and moral responsibility. Philosophical Psychology, 18, 561–584.

35.

Paulhus

D. L.

Carey

J. M.

(2011). The FAD-Plus: Measuring lay beliefs regarding free will and related constructs. Journal of Personality Assessment, 93, 96–104.

36.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891.

37.

Rakos

R. F.

Laurene

K. R.

Skala

Slane

(2008). Belief in free will: Measurement and conceptualization innovations. Behavior and Social Issues, 17, 20–39.

38.

Rockwood

N. J.

Hayes

A. F.

(2017, May). MLmed: An SPSS macro for multilevel mediation and conditional process analysis. Poster presented at the Annual Meeting of the Association of Psychological Science (APS), Boston, MA.

39.

Rucker

D. D.

Preacher

K. J.

Tormala

Z. L.

Petty

R. E.

(2011). Mediation analysis in social psychology: Current practices and new recommendations. Social and Personality Psychology Compass, 5, 359–371.

40.

Rudd

Vohs

K. D.

Aaker

(2012). Awe expands people’s perception of time, alters decision making, and enhances well-being. Psychological Science, 23, 1130–1136.

41.

Sarkissian

Chatterjee

De Brigard

Knobe

Nichols

Sirker

(2010). Is belief in free will a cultural universal? Mind & Language, 25, 346–358.

42.

Schönbrodt

F. D.

Perugini

(2013). At what sample size do correlations stabilize? Journal of Research in Personality, 47, 609–612.

43.

Seto

Hicks

J. A.

(2016). Disassociating the agent from the self: Undermining belief in free will diminishes true self-knowledge. Social Psychological and Personality Science, 7, 726–734.

44.

Shariff

A. F.

Greene

J. D.

Karremans

J. C.

Luguri

J. B.

Clark

C. J.

Schooler

J. W.

Baumeister

R. F.

Vohs

K. D.

(2014). Free will and punishment: A mechanistic view of human nature reduces retribution. Psychological Science, 25, 1563–1570.

45.

Shipp

A. J.

Edwards

J. R.

Lambert

L. S.

(2009). Conceptualization and measurement of temporal focus: The subjective experience of the past, present, and future. Organizational Behavior and Human Decision Processes, 110, 1–22.

46.

Snyder

C. R.

Sympson

S. C.

Ybasco

F. C.

Borders

T. F.

Babyak

M. A.

Higgins

R. L.

(1996). Development and validation of the State Hope Scale. Journal of Personality and Social Psychology, 2, 321–335.

47.

Spencer

S. J.

Zanna

M. P.

Fong

G. T.

(2005). Establishing a causal chain: Why experiments are often more effective than mediational analyses in examining psychological processes. Journal of Personality and Social Psychology, 89, 845–851.

48.

Stillman

T. F.

Baumeister

R. F.

Mele

A. R.

(2011). Free will in everyday life: Autobiographical accounts of free and unfree actions. Philosophical Psychology, 24, 381–394.

49.

Vohs

K. D.

Schooler

J. W.

(2008). The value of believing in free will: Encouraging a belief in determinism increases cheating. Psychological Science, 19, 49–54.

50.

Wegner

D. M.

(2004). Précis of the illusion of conscious will. Behavioral and Brain Sciences, 27, 649–659.

51.

Zhang

Yuan

K.-H.

(2018). Practical statistical power analysis using Webpower and R. ISDSA Press. https://webpower.psychstat.org