Mental rotation is a weak measure of people's propensity to visualise

In mental rotation (MR) tasks people can be asked to decide if pairs of objects depicted from different viewpoints are the same, or different. A common response strategy is to visualise one of the two objects rotating, until it is visualised from the same viewpoint as the other object. However, some people, Congenital Aphants, assert that they cannot visualise, and yet they perform similarly on MR tasks. This could mean that Congenital Aphants are mistaken about their inability to visualise. Alternatively, we reasoned that MR tasks might be an unreliable metric of people's propensity to rely on visualisation in MR tasks. In a sample of the general population, we had people report on their response strategies on a trial-by-trial basis. Neither people's overall propensity to visualise nor their propensity to visualise on different viewpoint trials was related to viewpoint-contingent changes in MR task performance. There was only a weak association between viewpoint-contingent changes in MR task performance and viewpoint-contingent changes in the proportion of visualising trials. Overall, our data suggest that MR task performance is a weak measure of people's propensity to visualise.