Date created: | Last Updated:

Category: Project

Description: The rubber hand illusion (RHI), in which participants report experiences of ownership of a fake hand, appears to demonstrate that the sense of ownership over one’s body can be disrupted by simple multisensory stimuli. In the RHI, ownership is typically quantified through subjective ratings of agreement with statements putatively about body ownership and such ratings are used to support an extensive literature on embodiment. It was recently shown that propensity to agree with these statements is correlated with trait phenomenological control (the ability to generate experience to meet expectancies arising from direct or indirect suggestion) and that existing methods of controlling for suggestion effects in the RHI are invalid. Consequently, ownership ratings in RHI can be considered to reflect implicit imaginative suggestion effects. This presents substantial problems for interpretation of RHI experiments. Here, we present the results of simulated experiments designed to mimic standard practice in the field using samples of real participants. In each experiment, the participant sample is differentially biased in selection for trait phenomenological control. We find that using samples comprised entirely of participants higher in trait phenomenological control almost guarantees that an experiment will provide evidence for the RHI. By contrast, experiments with samples comprised of only participants lower in trait phenomenological control will find evidence for the RHI only around half the time. It is therefore likely that experiments using the RHI are studying phenomenological control by proxy. Existing RHI experiments cannot be used to make inferences about the sense of ownership in humans generally.