Increasing prosocial behavior and decreasing selfishness in the lab and everyday life

The paper addresses how to increase prosocial behavior and reduce selfishness within naturally occurring relationships, focusing on couples. While humans uniquely engage in prosocial cooperation, they also exhibit selfish behavior, creating a central tension relevant to social interactions and conflicts. Laboratory behavioral games (e.g., the Dictator Game) standardize prosocial vs. selfish choices but face external validity challenges because such paradigms often involve decisions with strangers in artificial settings. The authors propose linking laboratory tests with real-life assessments to improve ecological validity. They target psychological flexibility—a malleable set of skills fostering awareness, openness, and values-consistent action—as a candidate mechanism to promote prosociality. They combine a micro-intervention to enhance psychological flexibility with ESM to capture daily prosocial acts. Hypotheses: (1) dyads receiving the micro-intervention would choose more prosocial outcomes in a behavioral game than dyads without the intervention; (2) dyads in which both partners received the intervention would show more prosocial outcomes than dyads with only one trained partner, and both would exceed dyads with no intervention (dose–response); (3) individuals receiving the intervention would show increases in prosocial behaviors across one week as measured by ESM.

Behavioral games, especially the Dictator Game, have been extensively used to study prosocial vs. selfish behavior, but findings with strangers may not generalize to real-life relationships due to differences in trust, exploitation concerns, and expectations of future interaction. Prior research has largely focused on fixed interpersonal factors (e.g., closeness, attraction, personality), game characteristics (e.g., payoffs), and contextual influences (e.g., mood, priming), or on biological interventions (e.g., oxytocin), which are outside everyday voluntary control. Few studies have experimentally trained skills aimed at increasing prosocial behavior. Some evidence suggests mindfulness and compassion training can increase prosocial orientations and behaviors toward strangers. Psychological flexibility is associated with health and social functioning and may help people act in line with values (e.g., being a supportive partner) even in the presence of internal or external barriers. ESM offers high ecological validity, reduces recall bias, and enables mechanism testing in daily contexts, supporting a multi-method approach linking lab and field.

Design: Two studies with couples (dyads). Preliminary study randomized dyads to receive a brief micro-intervention (approximately 15 minutes) targeting psychological flexibility or not, followed by a Dictator Game one week later. The main study randomized dyads into four groups: Group 1: both partners received the micro-intervention and ESM with exercises and questions about altruistic behavior; Group 2: one partner received the micro-intervention, both received ESM; Group 3: neither received the micro-intervention, both received ESM; Group 4: neither received the micro-intervention nor ESM. Group 2 tested dose–response; Group 4 controlled for ESM effects. All groups completed the Dictator Game one week post-intervention. Participants: Preliminary study recruited couples from local advertisements (text indicates dyads randomized micro-intervention n=10 vs. control n=6; “Subjects” section reports 8 couples; descriptive details primarily pertain to the main study). Main study recruited 494 individuals; 126 couples (225 individuals) randomized; 118 couples entered pre-assessment; after dropouts and exclusions, n=111 dyads (222 individuals) remained (114 females, 108 males; mean age 32 years, SD 12.24, range 18–75; relationship length mean 7.8 years, SD 9.04, range 0.5–44). Groups were balanced on age and relationship length. ESM compliance was high (96.4% completed >50% prompts); six individuals with <50% ESM completion were excluded from ESM analyses. Randomization and blinding: Random blocks with computer-generated sequences; stratified by age (>30 vs. ≤30). Allocation was communicated by an independent administrator. Participants were blinded to allocation; investigators were blinded to hypotheses. Procedure: Two lab visits (T1 and T2) 7 days apart. At T1, questionnaires were completed. Between T1 and T2, Groups 1–3 completed ESM on smartphones. At T2, all participants completed questionnaires and behavioral tests, including the Dictator Game. Intervention (micro-intervention): A 15-minute session training psychological flexibility. Steps: (1) identify current struggles/stressors; (2) clarify deeply valued people/activities (values); (3) experiential exercise to explore intersections of struggles and values; (4) practice present-moment awareness and acceptance to increase flexibility in pursuing values amid discomfort; (5) re-affirm values and plan values-consistent action. A take-home diary provided practice instructions (including present-moment awareness via audio), and participants were instructed to practice daily between T1 and T2. Measures:

• Prosocial behavior (lab): Dictator Game with dyad partners instead of strangers; each participant allocated 240 Swiss Francs worth of gift certificates between self and partner.
• Prosocial behavior (daily life via ESM): Smartphone prompts scheduled six times per day (approximately 07:00, 10:00, 13:00, 16:00, 19:00, 22:00) for one week between T1 and T2. Prosocial support to partner was queried five times per day (not during the first morning prompt). Items included: whether participants helped/supported someone since the last prompt and the degree of support specifically given to their partner (Likert scale 0 “not at all” to “very much”). Statistical analyses: Dictator Game amounts analyzed with multilevel models (participants nested in dyads), with intervention conditions modeled at dyad level (both/one/none) and at partner level (received vs. not received). Temporal ESM trajectories analyzed with three-level models (time within partners within dyads), with intervention (partner level) as between-subjects factor, testing linear and quadratic trends and interactions.

Preliminary study: One week after the intervention, participants who received the micro-intervention allocated on average 146 CHF (SD 49.9) to their partners vs. 120 CHF (SD 0.0) in controls, a 21.6% higher allocation (effect size d = 0.74). Main study (Dictator Game): Mean allocations by group: Group 1 (both partners intervention) 144.5 CHF (SE 6.2); Group 2 (one partner intervention) 133.8 CHF (SE 6.5); Group 3 (no intervention, ESM) 124.0 CHF (SE 6.72); Group 4 (no intervention, no ESM) 125.6 CHF (SE 6.99). Groups 3 and 4 did not differ and were combined for analyses. There was a significant linear trend across Group 1 > Group 2 > Groups 3&4 combined (p = 0.014; multilevel model). Collapsing across groups by partner-level intervention status showed a stronger effect: those who received the micro-intervention gave 141.3 CHF (SE 5.0) vs. 126.6 CHF (SE 4.2) for those without (p = 0.024; multilevel model). Response categories: Across all groups, 70.1% gave fairly (50/50), 20.1% generously, and 9.8% selfishly. More prosocial responses (fair + generous) occurred in Groups 1 and 2; Groups 3 and 4 had about three times more selfish responses. Among partner-level intervention groups: without intervention, 69% fair, 18% generous, 13% selfish; notably, 81% (17/21) of selfish responses came from those without the intervention vs. 19% (4/21) from those with the intervention. ESM (daily prosocial support): In Groups 1–3, Group 3 showed a decline over time, but the group-by-days interaction was not significant. Splitting by partner-level intervention, trajectories differed: those with the micro-intervention showed an initial slight decline followed by an increase toward the end of the week, whereas those without showed continued decline. The linear interaction was not significant (p = 0.167), but the quadratic interaction of time × intervention was significant (p = 0.014). Overall, across studies, the micro-intervention increased prosocial behaviors by 28% and decreased selfish behaviors by 35%, with evidence of dose–response effects.

Findings demonstrate that a very brief micro-intervention targeting psychological flexibility increases prosocial behavior and reduces selfishness among couples, both in a laboratory Dictator Game and in participants’ daily lives as captured by ESM. The dose–response pattern—highest prosociality when both partners were trained, intermediate when one partner was trained, and lowest when neither was trained—supports the intervention’s specific contribution. Compared to meta-analytic norms with strangers in the Dictator Game, couples in ongoing relationships showed markedly more prosocial allocations (about 70% fair and 20% generous), consistent with higher trust and expectations of future interaction. Importantly, the intervention further enhanced prosociality beyond these elevated baselines. The multi-method approach triangulated effects across standardized lab tasks and ecologically valid daily assessments. Practically, enhancing psychological flexibility appears to help individuals act in line with social values (e.g., supporting a partner) despite internal and external barriers, suggesting applicability to improving relationship dynamics and reducing social discord. The brevity and voluntary skill focus of the intervention indicate potential scalability to diverse social contexts.

A 15-minute psychological flexibility micro-intervention increased prosocial behavior and decreased selfishness in couples in both laboratory behavioral games and everyday life, with stronger effects when both partners were trained. The study links laboratory measures to ecologically valid daily behaviors, underscoring psychological flexibility as a viable, scalable target for promoting prosocial action. Future research should examine durability of effects, mechanisms mediating daily prosocial change, applications to other naturally occurring groups (families, teams, communities) and to interactions with strangers, and optimization of dosage and delivery formats to maximize impact.

(1) Although participants allocated real stakes via mutually exclusive gift certificates in the Dictator Game, some may have intended to share outside the game, potentially confounding allocations; no systematic evidence of this was found, but it cannot be fully ruled out. (2) ESM relies on self-report and reflection, which can introduce bias despite high ecological validity and reduced recall error. Additional limitations include minor inconsistencies in prompt frequency reporting and potential generalizability constraints beyond couples in a specific cultural context.