Diverging implicit measurement of sense of agency using interval estimation and Libet clock

Besides relying on explicit agency ratings, research concerning the sense of agency seeks to employ implicit measures. One such measure is intentional binding, i.e., the shortening of the perceived temporal interval between an action and its consequence. Multiple paradigms have been used to measure intentional binding. Our aim was to compare two common paradigms within the same participants: interval estimation and the Libet clock paradigm. As both are supposed to measure intentional binding, we hypothesized the two measures correlated positively. In two experiments, we used Bayes factors as a stopping criterion for data collection in a correlational study design resulting in valid samples of 46 and 36 healthy adult students. Despite good reliability, Bayes factors showed moderate evidence for intentional binding from the two paradigms to be uncorrelated. Implications for future research of sense of agency and the necessity for nomological validity of its measures are discussed.