Co-representation of human-generated actions vs. machine-generated actions: Impact on our sense of we-agency?

摘要

Many studies suggest that individuals are not able to build a sense of we-agency during joint actions with automated artificial systems. We sought to examine whether or not this lack of sense of control was linked with individuals' inability to represent automaton-generated action into their own cognitive system. Indeed, during human interactions, we automatically represent our partner's actions in our own sensorimotor system. This might sustain our capacity to build a sense of we-agency over our partner-generated actions. However, to our knowledge, no studies have investigated the potential relation between our ability to use our sensorimotor system to represent a partner's action and we-agency. Our approach consisted in performing a pilot study coupling together a Simon target detection task wherein RTs served as an index of action co-representation and an implicit measure of one's sense of agency as indicated by intentional binding phenomenon. The preliminary observations suggested that individuals could represent other-generated action and have a sense of agency over these actions provided that their partner was another human and not an artificial automated system.