Two minds, one body: neural correlates of processing of We-information during shared control over a humanoid robot

When speaking about the self, we typically think about ‘I’ or ‘Me’. However, our relations with other people are also a critically important aspect of our self, forming our ‘We’. In the current study, we investigated processing of information about the We during joint control over a single body. While naturally such a case occurs among the conjoint twins, we reconstructed such a situation by allowing two people to share control over a single humanoid robot. Using EEG, we investigated which neurocognitive processes are affected by providing We-information, i.e. information that conveys knowledge about properties at the group-level, while being uninformative about the individuals. We found that processing of We-information engages a cascade of processes, as indexed by various event-related potentials (ERPs), involving perception (P1, N1 and P2p), action selection (N2), working memory (P3b) and response preparation (contingent negative variation, CNV). We-information also modulates perception and processing of the subsequent information about the robot's target position, as indicated by modulation of similar ERPs (P1, N2 and P3b). Moreover, we found that We-information can affect dynamic allocation of covert attention towards the side of the partner controlling the other side of the robot. Overall, our results show that processing of information about the We can have a widespread influence on cognition, similarly to processing information about the self. This article is part of the discussion meeting issue ‘From being to identity: analysis and synthesis of the self’.