TMS-induced inhibition of the left premotor cortex modulates illusory social perception

Abstract

Prior experiences help us to navigate social interactions. Communicative actions from one person are used to predict another person's response. However, in some cases, these predictions can outweigh the processing of sensory information and lead to illusory social perception such as believing to see two people interact, although in real, there is only one person present. This phenomenon was referred to as seeing a Bayesian ghost. In this pre-registered study, we used a two point-light agents design. One agent was well-recognizable and performed either a communicative or individual gesture. The other agent either was blended into a cluster of noise dots and responded to this gesture or it was entirely replaced by noise dots. The participants had to indicate the presence or absence of the masked agent. Before the task, we applied either inhibitory non-invasive brain stimulation over the left premotor cortex (i.e., real TMS) or sham TMS as control condition. As expected, participants had more false alarms (i.e., Bayesian ghosts) in the communicative than individual condition in the sham TMS session and this difference between conditions vanished in the real TMS session. In contrast to our hypothesis, the general level of false alarms increased (rather than decreased) in the real TMS session for both conditions. These findings confirm that the premotor cortex is causally linked to the illusory social perception of a Bayesian ghost, which is likely due to disrupted action predictions and of clinical significance.