What Makes Action and Outcome Temporally Close to Each Other: A Systematic Review and Meta-Analysis of Temporal Binding

Temporal binding refers to the subjective compression of the temporal interval between a voluntary action and its external sensory consequences. While empirical evidence and theoretical accounts have indicated the potential linkage between temporal binding and action outcome prediction mechanisms, several questions regarding the underlying processes and the fundamental nature of temporal binding remain unanswered. Based on the sophisticated classification of predictive processes proposed by Hughes et al. (2013), we conducted a systematic, quantitative review of the binding effect as measured with two representative procedures, i.e., Libet clock procedure and interval estimation procedure. Although both procedures were designed to measure the same phenomenon, we revealed a larger effect size and higher sensitivity to perceptual moderators in binding observed with the clock procedure than with the interval estimation. Moreover, in the former, we observed different characteristics for the two perceptual shifts that comprise temporal binding. Action shifts depended more on whether one can control outcome onsets with voluntary actions. In contrast, outcome shifts depended more on the degree to which participants could predict, rather than control, the action outcome onset. These results indicate that action shift occurs based on the activation of learned action-outcome associations by planning and executing actions, while outcome shift occurs based on comparing predicted and observed outcomes. By understanding the nature of each experimental procedure and each shift, future research can use optimal methods depending on the goal. We discuss, as an example, the implications for the underlying disorders of agency in schizophrenia.