Counterfactual structure and learning from experience in negotiations

Reflecting on the past is often a critical ingredient for successful learning. The current research investigated how counterfactual thinking, reflecting on how prior experiences might have been different, motivates effective learning from these previous experiences. Specifically, we explored how the structure of counterfactual reflection – their additive (‘‘If only I had”) versus subtractive (‘‘If only I had not”) nature – influences performance in dyadic-level strategic interactions. Building on the functionalist account of counterfactuals, we found across two experiments that generating additive counterfactuals about a previous negotiation produced an advantage for negotiators over their previous performance compared to subtractive counterfactuals, both in terms of obtaining value for oneself and conceiving creative agreements. Additive counterfactuals enabled negotiators to more effectively extract lessons from past experiences to improve their current negotiation performance. ! 2009 Elsevier Inc. All rights reserved. Learning from experience often involves determining the causes of performance shortcomings to identify lessons to improve performance. A useful tool for establishing causal links between antecedents and outcomes is counterfactual thinking, or considerations of ‘‘what might have been” (Kahneman &Miller, 1986; Kray, Galinsky, & Wong, 2006; Markman & McMullen, 2003; Wells & Gavanski, 1989). Counterfactual thinking has proven to be a fundamental part of improving subsequent performance (Markman, Gavanski, Sherman, & McMullen, 1993). Whether it is students aiming to improve their scores on academic exams (Roese, 1994) or pilots learning from disasters barely averted (Morris & Moore, 2000), the generation of counterfactuals improves performance by helping to specify the necessary conditions to avoid repeating previous errors. One aspect of counterfactuals that determines their impact is the direction of the counterfactual, or whether a better versus worse possible world is imagined. Comparisons to worse possible worlds (downward counterfactuals) can help people feel better; whereas comparisons to better possible worlds (upward counterfactuals) can help to improve future performance (though at a temporary cost to affect). Although previous research has established that the generation of upward counterfactual thoughts about a past negotiation increases time devoted to preparing for subsequent negotiations (Galinsky, Seiden, Kim, & Medvec, 2002), surprisingly little attention has been paid to the effect of counterfactual thought on actual learning (i.e. performance gains) from past strategic interactions. The current research holds constant the direction of counterfactuals and examines how counterfactual structure (additive versus subtractive) impacts the learning process in strategic interactions. Whereas additive counterfactuals introduce new antecedent elements in constructing an alternate reality (e.g., ‘‘If only I had studied X topic, I would have gotten a better grade”), subtractive counterfactuals remove antecedent elements (e.g., ‘‘If only I had not studied this obscure topic, I would have gotten a better grade”) (Roese, Hur, & Pennington, 1999; Roese & Olson, 1993). In Roese (1994) influential research on the functions of counterfactuals, counterfactual direction and structure orthogonally impacted learning: both upward and additive counterfactuals improved performance on subsequent anagram tasks. We hypothesize that the generation of additive counterfactuals produces an advantage in mixed-motive interactions. By identifying alternative actions that would have resulted in success, additive counterfactuals provide a specific script for future action (Johnson & Sherman, 1990; Roese, 1994; Roese & Olson, 1993). In contrast, subtractive counterfactuals simply remove an option from consideration, thereby leaving negotiators with a more poorly specified roadmap for future action. Additive counterfactuals are also more creative than subtractive counterfactuals (Markman, Lindberg, Kray, & Galinsky, 2007; Roese, 1994). Whereas subtractive counterfactuals simply remove one response option from consideration, additive counterfactuals necessitate going beyond the original premise set to generate novel options not previously considered. Because negotiating effectively often involves a creative 0022-1031/$ see front matter ! 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jesp.2009.03.008 * Corresponding author. E-mail address: [email protected] (L.J. Kray). Journal of Experimental Social Psychology 45 (2009) 979–982