Demand characteristics confound asynchronous control conditions in indirect measures of the rubber hand illusion.
Reports of experiences of ownership of a fake hand following synchronous tactile and visual stimulation of fake and real hands have been attributed to multisensory integration mechanisms (the rubber hand ‘illusion’). However, it has been shown that the subjective reports expected, namely stronger experiences in the synchronous than asynchronous condition, and for ‘illusion’ rather than ‘control’ statements, are clear to subjects; thus, the reports may reflect response to demand characteristics. Subjective report is often accompanied by so-called ‘objective’ or ‘implicit’ measures. Like subjective report, an asynchronous control condition is typically employed for these measures. Here we report tests of expectancies for synchronous ‘illusion’ and asynchronous ‘control’ conditions for two indirect measures of rubber hand effects across two pre-registered studies (n = 140 and n = 45): proprioceptive drift (a change in perceived hand location) and skin conductance response (SCR; a measure of physiological arousal). Participants expectancies for the synchronous condition measures were greater than for asynchronous ‘control’ condition in both studies. Previous work has shown that subjects can voluntarily produce alterations in both subjective experience of body location and in SCR. Thus, indirect measures of rubber hand effects may reflect compliance, bias and phenomenological control in response to demand characteristics, just as for subjective measures, and the asynchronous control condition is in general invalid for controlling demand characteristics in rubber hand procedures. Valid control measures are required to support claims of a role of multi-sensory integration for both direct and indirect measures of rubber hand effects.