Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies.

Given the powerful implications of relationship quality for health and well-being, a central mission of relationship science is explaining why some romantic relationships thrive more than others. This large-scale project used machine learning (i.e., Random Forests) to 1) quantify the extent to which relationship quality is predictable and 2) identify which constructs reliably predict relationship quality. Across 43 dyadic longitudinal datasets from 29 laboratories, the top relationship-specific predictors of relationship quality were perceived-partner commitment, appreciation, sexual satisfaction, perceived-partner satisfaction, and conflict. The top individual-difference predictors were life satisfaction, negative affect, depression, attachment avoidance, and attachment anxiety. Overall, relationship-specific variables predicted up to 45% of variance at baseline, and up to 18% of variance at the end of each study. Individual differences also performed well (21% and 12%, respectively). Actor-reported variables (i.e., own relationship-specific and individual-difference variables) predicted two to four times more variance than partner-reported variables (i.e., the partner's ratings on those variables). Importantly, individual differences and partner reports had no predictive effects beyond actor-reported relationship-specific variables alone. These findings imply that the sum of all individual differences and partner experiences exert their influence on relationship quality via a person's own relationship-specific experiences, and effects due to moderation by individual differences and moderation by partner-reports may be quite small. Finally, relationship-quality change (i.e., increases or decreases in relationship quality over the course of a study) was largely unpredictable from any combination of self-report variables. This collective effort should guide future models of relationships.

Interindividual-intergroup discontinuity in the domain of correspondent outcomes: the roles of relativistic concern, perceived categorization, and the doctrine of mutual assured destruction.

Most prior research on the tendency for groups to be less cooperative than individuals (the interindividual-intergroup discontinuity effect) has used the Prisoner's Dilemma Game (PDG). Experiment 1 examined the discontinuity effect with 3 additional matrices: Chicken, Leader, and Battle of the Sexes (BOS). Unlike the PDG, these matrices are characterized by correspondence of outcomes. The discontinuity effect was significant for the PDG and Chicken matrices only. With the BOS and Leader matrices, both individuals and groups pursued outcome maximization through coordinated turn taking. Despite the lesser competitiveness, sets of interacting participants in the BOS and Leader conditions did perceive that they were 2 groups. Experiment 2 examined the discontinuity effect in 2 Chicken matrices with varying outcomes associated with mutual competition. Consistent with the doctrine of mutual assured destruction, the discontinuity effect was eliminated for the matrix in which mutual competition was associated with very low outcomes. Although concern for relative in-group standing gave rise to intergroup competition even in the domain of correspondent outcomes, such concern was constrained to the extent that it interfered with outcome maximization.

Reduction of interindividual-intergroup discontinuity: the role of leader accountability and proneness to guilt.

Two experiments contrasted interactions between group leaders with interactions between individuals in a mixed-motive setting. Consistent with the idea that being accountable to the in-group implies normative pressure to benefit the in-group, Experiment 1 found that accountable leaders were more competitive than individuals. Consistent with the idea that being unaccountable to the in-group implies normative pressure to be cooperative and that high guilt proneness provides motivation to be moral, Experiment 2 found that when guilt proneness was high, unaccountable leaders were less competitive than accountable leaders and did not differ significantly from individuals. In other words, the robust interindividual-intergroup discontinuity effect was eliminated when groups had unaccountable leaders who were high in guilt proneness.

Interindividual-intergroup discontinuity as a function of trust and categorization: the paradox of expected cooperation.

Two experiments used a 3-choice variation of the prisoner's dilemma game to explore the paradoxical implications of expected cooperation of other groups and individuals for competitiveness and cooperativeness. Experiment 1 found that an experimental manipulation of opponent trustworthiness influenced the tendency of both groups and individuals to cooperate but had no significant effect on the tendency of groups and individuals to compete--possibly because of the perceived unfairness of competing with a trustworthy other. Experiment 2 found that an experimental manipulation of categorization increased the tendency of same-category groups and individuals to cooperate (as in Experiment 1) and also increased the tendency of same-category groups, but not individuals, to compete (unlike in Experiment 1). It was further found that the tendency of same-category groups to compete more than same-category individuals was relatively stronger for participants high in guilt proneness.