New #psyarxiv preprint: ‘Demand characteristics confound the rubber hand illusion’ psyarxiv.com/mpncv. Methods to control for demand characteristics in the illusion are not fit for purpose. Illusion reports are likely to reflect suggestion and compliance effects. (1/7) 
In rubber hand illusion studies, demand characteristics are usually controlled for by contrasting agreement scores for ‘illusion’ statements (synchronous condition referred touch and ownership experience) and ‘control’ statements (e.g., hallucinations). (2/7)
Higher scores for ‘illusion’ than ‘control’ measures are considered evidence that ‘illusion’ scores are not attributable to suggestion. A critical untested assumption of this method is that participant expectancies are closely matched across illusion and control conditions. (3/7)
Here, expectancies for rubber hand illusion and classic control statements were measured after participants watched a video showing the illusion induction procedure. (4/7)
Mean expectancy ratings were higher for ‘illusion’ than for ‘control’ statements. Differences in agreement scores in rubber hand illusion studies may merely reflect differing demand characteristics (5/7)
RHI control methods are confounded by demand characteristics. The rubber hand illusion may be an implicit imaginative suggestion effect driven by top-down phenomenological control (see psyarxiv.com/82jav). (6/7)
Valid control measures will be required to establish evidence of rubber hand illusion experience which cannot be easily attributed to demand characteristics (some suggestions are made). (7/7)
