Revisiting the temporal pattern of regret in action and inaction : replication and extension of Gilovich and Medvec (1994)

Abstract

Gilovich and Medvec (1994) investigated the temporal pattern of regret, in which participants perceived or experienced stronger regret with action in the short-term, but stronger regret with inaction in the long term. Following mixed and null findings, it is important to directly replicate the studies. In a pre-registered experiment with a sample from Amazon Mechanical Turk (Post-exclusion N = 988, Pre-exclusion N = 1017), we replicated Studies 1 (V = 0.50, 95% CI [0.27, 0.70], inaction effect), 3 (short-term V = 0.53, 95% CI [0.35, 0.70], action-effect; long-term V = 0.28, 95% CI [0.05, 0.48], inaction-effect), 4 (short-term V = 0.53, 95% CI [0.24, 0.76], action-effect; long-term V = 0.24, 95% CI [0.00, 0.52], inaction-effect) and 5 (short-term V = 0.06, 95% CI [0.00, 0.44]; long-term V = 0.56, 95% CI [0.25, 0.81]) from Gilovich and Medvec (1994). We found support for action-effect in the short-term, and inaction-effect in the long-term in Study 1 (V = 0.25, 95% CI [0.17, 0.34], inaction effect), Study 3 (short-term V = 0.23, 95% CI [0.14, 0.31], action-effect; long-term V = 0.15, 95% CI [0.06, 0.24], inaction-effect), and Study 4 (short-term V = 0.24, 95% CI [0.12, 0.36], action-effect; long-term V = 0.10, 95% CI [0.01, 0.23]), but with smaller effect sizes. We failed to find support for such a pattern in Study 5 (short-term V = 0.05, 95% CI [0.00, 0.13]; long-term V = 0.04, 95% CI [0.00, 0.13]). The discrepancy in findings may be due to differences in meanings of “action” and “inaction” between Study 5 (doing something vs not doing something) and Study 3 or 4 (change vs no change), or differences in perceiving other emotions and actual personal experiences, or certain scenario constraints of generality. Extending the replication, we found support for stronger responsibility for action compared to inaction both in the short-term and the long-term. More pre-registered replications and follow-up work are necessary to resolve the inconsistencies in findings. Supplementary, materials, raw data, and analysis files/code are available here: https://osf.io/7m3q2/