The effect of future self-continuity on intertemporal decision making: a mediated moderating model

The study examines how perceived continuity between the present and future self (future self-continuity) shapes intertemporal decision-making, particularly the time discount rate, and whether this effect depends on self-concept clarity and is transmitted through consideration of future outcomes. Intertemporal choice often involves trading off smaller-sooner versus larger-later rewards, with time discounting reflecting reduced subjective value of delayed outcomes. Prior work suggests higher future self-continuity lowers discounting, fostering patience. However, boundary conditions and mechanisms are less clear. The authors propose: H1) Future self-continuity predicts a lower time discount rate; H2) Self-concept clarity moderates the impact of future self-continuity on time discounting; H3) Consideration of future outcomes mediates the moderating effect of self-concept clarity on the relationship between future self-continuity and intertemporal decision-making. The work aims to clarify when and how future self-continuity promotes patient choices, with implications for economic, health, and prosocial behaviors.

The introduction surveys research on intertemporal choice and time discounting, noting that larger discount rates indicate a preference for immediate rewards and are often seen as short-sighted. Future self-continuity—perceived similarity, vividness, and positivity of one’s future self—has been linked to greater patience, saving, prosocial behavior, and healthful actions. Mechanistically, greater psychological connectedness reduces perceived temporal and psychological distance, encouraging long-term planning and reduced discounting. The authors review self-concept clarity, defined as a clear, consistent, and stable self-view, which precedes and supports self-continuity development and self-control. Theories of possible selves and multiple self-concepts suggest that clarity can align present and future goals, enabling rational delayed choices. The consideration of future consequences (future outcome consideration) shapes how individuals weigh long-term outcomes in decisions; higher consideration correlates with saving, health behaviors, and lower discounting, while low consideration is associated with immediate gratification. These threads motivate a model where self-concept clarity moderates the link between future self-continuity and outcomes partly via enhanced consideration of future consequences.

Study 1 (Questionnaire, correlational): - Participants: 502 invited via Questionnaire Star; 132 failed attention checks; final N=370 (264 female, 106 male), age 18–35 (M=19.54, SD=3.02). Ethics approval and informed consent obtained. - Measures: - Future Self-Continuity Questionnaire (FSCQ; Sokol & Serper, 2020): 10 items across similarity (4), vividness (3), positivity (3), 7-point Likert. Reliability: total α=0.900; subscales α=0.854 (similarity), 0.722 (vividness), 0.893 (positivity). - Self-Concept Clarity Scale (SCCS; Campbell et al., 1996): 12 items, mostly reverse-scored, 7-point scale. α=0.841. - Intertemporal decision-making: Monetary-Choice Questionnaire (27 choices SIR vs LDR; Kirby et al., 1999) with real incentives (1% chance; immediate or delayed 1 week for chosen trial). Time discount rate computed from choices. - Objective Socio-Economic Status (OSES): per capita monthly household income, 9-point scale. - Procedure and analysis: Online completion (~10 min). Controlled for common method variance (Harman’s single-factor test; largest factor 18.466%). Descriptives and correlations in SPSS 25. Structural equation modeling (SEM) in Mplus 8.3: FSC as latent independent variable, time discount rate (TDR) as latent dependent variable, SCC as latent moderator; controls: age, gender, OSES. Moderation tested using latent interaction with Bayesian estimation due to convergence advantages (PSR≈1.007 at 2000 iterations). Study 2 (Experiment, moderated mediation): - Participants: Power analysis (d=0.4, α=0.05) indicated ≥200 for 80% power. Recruited 262 undergraduates; 28 failed attention checks; final N=234 (89 males; M age=18.71, SD=1.40). Right-handed, normal/corrected vision. Randomly assigned to high vs low FSC groups (n=117 each). Ethics approval and consent. - Design: 2 (FSC: high vs low; between-subjects). Independent variable: manipulated FSC; Moderator: SCC (latent); Mediator: Future Outcome Consideration (FOC; future dimension); Dependent: TDR. Controls: gender, age, OSES. - Materials and tasks: - FSC manipulation: Reading vignettes implying stability (high FSC) vs substantial change/fragmentation (low FSC) in traits/values over time, followed by a writing task listing similarities (high) or differences (low) between current self and self 10 years later (~50 words). Manipulation check used state-framed FSCQ (α total=0.906; subscales α=0.851, 0.771, 0.855). - FOC scale: Future dimension of Consideration of Future Consequences (5 items), 7-point scale, state framing. α=0.719. - SCCS: same as Study 1 (α=0.834). - Monetary-Choice Questionnaire: 27 computerized trials (E-Prime 2.0). Trial structure: fixation 500 ms; options displayed until response (left=SIR via F key; right=LDR via J key); chosen option highlighted for 1000 ms; inter-trial interval 800–1000 ms. Real incentive scheme analogous to Study 1. - OSES as in Study 1. - Procedure: Completed SCCS online, then manipulation and state measures (FSCQ, FOC), then intertemporal choice task. Practice trials ensured comprehension. - Analysis: SPSS 25 for t-tests, descriptives, correlations. Mplus 8.3 for SEM. Moderation and moderated mediation tested using latent interaction/product indicator methods with Bayesian estimation (PSR≈1.006–1.007 at 2000 iterations). Measurement models used item parceling/project balance where applicable. Controls (gender, age, OSES) included throughout.

Study 1: - Correlations: FSC negatively correlated with TDR (r = -0.163, p = 0.002) and positively with SCC (r = 0.102, p = 0.050). SCC–TDR correlation not significant (r = 0.053, p = 0.307). - Main effect SEM: Good fit; FSC significantly negatively predicted TDR (standardized b ≈ -0.154 to -0.217 across models; e.g., b = -0.154, p = 0.013, 95% CI [-0.277, -0.032]). - Moderation (Bayesian latent interaction): FSC → TDR negative (b = -0.247, p_total = 0.002, 95% CI [-0.399, -0.104]); SCC → TDR positive (b = 0.162, p_total = 0.022, 95% CI [0.002, 0.323]); interaction FSC×SCC negative (b = -0.232, p_total = 0.007, 95% CI [-0.427, -0.048]). Simple slopes: low SCC (-1 SD) b = -0.076, p_conditional = 0.219; high SCC (+1 SD) b = -0.418, p_conditional < 0.001. Study 2: - Manipulation check: High-FSC group > Low-FSC on FSCQ (M ± SD = 5.113 ± 0.690 vs 3.604 ± 0.629), t(232) = -17.479, p < 0.001, d = 2.286. - Correlations: FSC negatively with TDR (r = -0.150, p = 0.022); FSC positively with FOC (r = 0.312, p < 0.001). FOC negatively with TDR (r = -0.270, p < 0.001) and with SCC (r = -0.375, p < 0.001). FSC–SCC not significant (r = -0.048, p = 0.463). SCC–TDR not significant (r = 0.086, p = 0.189). - Main effect: FSC negatively predicted TDR (b = -0.154, p = 0.013, 95% CI [-0.277, -0.032]). - Moderation (Bayesian): FSC → TDR negative (b = -0.150, p_one-tailed < 0.009); SCC → TDR ns (b = 0.039, p_one-tailed = 0.326); interaction FSC×SCC negative (b = -0.202, p_one-tailed = 0.007). Simple slopes: low SCC (-1 SD) b = -0.016, p_one-tailed = 0.428; high SCC (+1 SD) b = -0.317, p < 0.001. - Moderated mediation: Direct FSC → TDR ns when mediator included (b = -0.056, p_one-tailed = 0.223). FSC → FOC positive (b = 0.263, p < 0.001). FOC → TDR negative (b = -0.367, p_one-tailed = 0.003). SCC moderates FSC → FOC (b = 0.209, p_one-tailed < 0.001); SCC does not significantly moderate FSC → TDR in this model (b = -0.122, p_one-tailed = 0.089). Indirect moderated effect via FOC significant at high SCC (+1 SD; Effect = -0.161, p = 0.003, 95% CI [-0.290, -0.020]) but not at low SCC (-1 SD; Effect = -0.030, p = 0.123). Overall moderating effect of SCC on FSC → TDR = -0.202, with indirect component via FOC ≈ -0.080 (≈39.6% of total moderation).

The findings support that perceiving a strong connection with one’s future self reduces temporal discounting, promoting patience in intertemporal choices. This addresses H1 and aligns with theories that reduced psychological distance between present and future selves enhances willingness to sacrifice immediate rewards. Critically, self-concept clarity emerged as a boundary condition (H2): only individuals with clearer, more stable self-views translated future self-continuity into lower discounting. This suggests that clarity facilitates aligning current behavior with long-term goals. Mechanistically, consideration of future outcomes helps explain how and when continuity matters (H3). Higher future self-continuity increased attention to long-term consequences, which in turn lowered discount rates; this pathway was stronger at higher self-concept clarity, fully accounting for the moderation when included in the model. The results therefore integrate identity continuity, self-structure clarity, and future-oriented cognition into a coherent account of intertemporal decision-making, with implications for interventions that bolster future self vividness and self-concept clarity to reduce myopic choices in financial, health, and prosocial domains.

Across a survey-based study and an experiment, future self-continuity consistently predicted lower time discount rates, indicating greater willingness to choose delayed, larger rewards. Self-concept clarity moderated this relationship: the beneficial effect of future self-continuity on patience appeared robust among individuals high in clarity but not among those low in clarity. Moreover, consideration of future outcomes mediated the moderating effect of self-concept clarity, such that greater continuity increased focus on long-term consequences, lowering discounting especially when self-concept clarity was high. These contributions clarify when and how identity-based processes shape intertemporal choice. Future research should extend beyond monetary paradigms to non-monetary domains and incorporate response time and process measures to better capture cognitive mechanisms. Interventions that enhance future self vividness and improve self-concept clarity may effectively reduce temporal discounting.

- Response dynamics: The experimental focus was on choice rates without time limits; individual differences in response time may have influenced outcomes. Future work should model or control response times. - Domain specificity: The paradigm centered on monetary choices; generalizability to non-monetary intertemporal decisions (e.g., health, environmental, academic) remains to be tested.