What Could Have Been or What Will Be: A Test of the Functional Theory of Counterfactual Thinking

The functional theory of counterfactual thinking (Epstude & Roese, 2008) proposes that considering how past events might have been different prepares us for how the future might be. Previous research provides ambiguous support: Counterfactual thoughts contain less controllable features than future thoughts (e.g. Ferrante, Girroto, Straga, & Walsh, 2013) and therefore cannot serve a preparatory function. In this study, we used a novel experimental stimulus to compare a narrow and diffuse time period counterfactual with future thinking. In prior experimental designs, when participants were expected to think counterfactually, the time period they were being asked to consider was ambiguous. Thus, the specificity of the targeted time period was confounded with temporal direction (counterfactual thinking versus future thinking). We removed this confound to see if previous results would replicate. They did replicate: Counterfactual thoughts contained less controllable features relative to future thoughts. Therefore, we concluded that the targeted time period being confounded with temporal direction did not contribute to results reported in previous literature. We also tested the controllability (controllable vs uncontrollable) and locus of causation (internal vs external) confound identified by Roese, Smallman, and Epstude (2017). We found no difference in the locus of causation of counterfactual and future thinking when considering all thoughts produced, but less internally-oriented features when thinking counterfactually relative to the future when considering only the first thought produced (as per Kahneman & Tversky, 1982).