Implied motion guides neither gaze following nor intention attribution

The "gaze beam hypothesis (GBH)" of gaze following posits that the others eyes emit imaginary beams of moving energy travelling to the others object of attention (1), drawing the observers attention to the same object. This idea was initially supported by behavioral experiments showing a motion aftereffect (MAE), indicated by longer reaction times in detecting motion direction after viewing a cartoon face looking at an object in the same direction (2). However, this effect could also be expected if the observer used gaze direction to assume an intentional link between the looker and the object, envisioning directed actions toward the latter (3). To critically compare the two hypotheses, we tested whether an MAE could be induced by having human subjects detect motion direction after viewing various cue images, designed to differentiate the explanatory power of the two. Cues either suggested a connection between an agent and an object through the agents gaze or an object-oriented intention by the presence of equipment in the agents hand, without the agent directly looking at the object. Using Bayesian statistics, our findings provided strong evidence against both hypotheses at the population level, as reaction time modulations did not align with the MAE, leading us to reject motion adaptation as the underlying mechanism for gaze following but also intention attribution. As on an individual level we observed highly diverse effects, with some compatible with one or the other hypothesis, we assume that individual subjects may resort to different perceptual strategies based on different scene interpretations.