To act or not to act? Implicit and explicit measures of the valence of the sense of agency

Cleeremans and Tallon-Baudry (2022) proposed that the core function of consciousness might be to enable agents to value their experiences. From this perspective, every conscious experience is affectively valenced, albeit to varying degrees. Interestingly, acting itself is associated with a specific phenomenology referred to as “sense of agency” (SoA, Pacherie 2008). Thus, if every conscious experience is valenced, and if performing an action is accompanied by a conscious SoA, then SoA should itself be valenced. Sixty participants completed an Affective Misattribution Procedure (Payne & Lundberg, 2014) in which we manipulated two factors: intentionality (voluntary vs. passive key selection) and causality (outcome matching vs. not matching the selection). We expected the SoA to be higher in the voluntary/matching blocks than in the passive/non-matching blocks. Following the SoA manipulation, participants’ task was to judge the valence of a priori neutral Chinese pictographs. We hypothesized that higher SoA should be more positively valenced, and therefore implictly influencing the rating of the pictographs toward positive valences. While explicit reports could corroborate our hypothesis, we did not observe evidence of an influence of the SoA manipulation on the valuation of pictographs (i.e., implicit measure). However, exploratory analyses highlighted a high variability of the effect suggesting that SoA could be differently valenced from one individual to another rather than not valenced at all. Future studies should explore what produces this difference in the experience of SoA.