Making predictions from kinematics and causal intuitions: a dissociation between judged causality and imagined locations

Abstract

To mentally extrapolate the trajectory of moving objects disappearing from sight, different sources of information can be exploited. Among them, the memory of its last visible position and movement, its inferred movement through time, and the general understanding of the causal relations between objects in the scene are particularly important. It is often assumed that these cues are integrated into unified analog mental representations. To investigate the mental representation of imagined movements and their relation with our intuitions about the nature of the physical events, we used a novel experimental paradigm in which a causal attribution task was combined with a motion prediction task. Participants were instructed to imagine the trajectory of a moving object disappearing behind a screen and to estimate the degree to which the movement was launched by another moving object. To test the impact of causality, we modified a launching event by introducing either a temporal delay or a spatial interval between the movements. Compared to a situation in which the target object remained visible through its entire trajectory, participants were unable to precisely predict the actual position of the target when it was occluded. Futhermore, their predictions were misaligned with their judgment of causality: participants predicted the actual position of invisible moving objects better for events that they judged less causally correct than for events they judged more causally correct. To see if some other high-level physical representations could affect the prediction, we tilted the slope of the track such that the balls would either roll downward or upward, resulting in no obvious difference from the horizontal track condition. Overall, our data suggest that kinematics and cognitive parameters of imagined dynamical events do not merge into a unitary mental model simulating actual states of the world.