Playing it safe: Negotiators avoid uncertainty and reach safer, but less integrative agreements

Social-interactive conflict management processes inherently involve uncertainties about the exact outcomes that agreements may yield. However, two literature reviews show that irrespective of the inherent uncertainty of most real-world social decision-making contexts, experimental negotiation studies have traditionally relied on certain outcome scenarios so that it remains unknown whether and how outcome uncertainty affects interactive logrolling and therefore agreement quality. In contrast, we propose and test two competing hypotheses on how outcome uncertainty may affect social-interactive joint decision-making. On the one hand, the Value-Insight Hypothesis suggests that focusing on the value component of uncertain outcomes highlights priority differences between conflict parties, thereby facilitating parties' interactive exploration of integrative priority-based trade-offs. On the other hand, the Uncertainty- Avoidance Hypothesis argues that focusing on the risk component of uncertain outcomes leads to mutual uncertainty-avoidant trade-offs that hinder parties to explore integrative priority-based trade-offs. We introduce novel social decision-making paradigms to systematically test these competing hypotheses in a causal chain design. Specifically, across three pre-registered (N = 502) interactive negotiation studies, we find that in line with the Uncertainty-Avoidance Hypothesis parties' sensitivity to explore priority-differences is reduced under uncertainty. Across different operationalizations of outcome uncertainty parties reach less integrative agreements for uncertain compared to certain outcomes. In fact, parties reached more compromising agreements under uncertainty irrespective of varying levels and ranges of risks (Study 2). Beyond uncertainty-avoidance, parties reached lower-quality agreements under uncertainty even in contexts in which no mutual trade-offs allowed them to reduce or avoid uncertainty (Study 3).