The time course of proprioceptive drift in the rubber hand illusion

In the well-known rubber hand illusion (RHI, Botvinick Cohen, Nature, 1998), synchronous tactile stimulation of a subject's invisible hand and a visible rubber hand (placed at a fixed lateral distance) induces a drift of perceived location of the real hand towards the rubber hand. The purpose of this study was to investigate the spatio-temporal characteristics of this proprioceptive drift. To this end, we measured the perceived position of the participants' index finger in relation to a small visible probe using an adaptive two-staircase method and a forced-choice task. We determined perceived finger location by fitting a psychometric function to the responses. Prior to introducing the rubber hand at a distance of 17cm from the real hand, we determined the perceived finger location in darkness. The time course of the RHI was then determined in three phases: pre-test while only looking at the rubber hand, prolonged synchronous tactile stimulation, and post-test again without touch but with visible rubber hand. The synchronous stimulation was fully controlled using two PHANToM force-feedback devices. The perceived finger location immediately shifted 1.4cm towards the rubber hand when in view. This shift rose to an average of 6.3cm after 8min of tactile stimulation. The distribution of responses indicates that the proprioceptive drift is truly gradual. As previous work suggests (cf.Holmes et al., PP, 2006; Tsakiris Haggard, JEP, 2005), these findings show that the RHI involves both immediate effects that result from multisensory integration as well as gradual recalibration effects with a time constant of several minutes. After 5mins of post-test, a drift of 4.9cm remained, showing that recalibration produces an after-effect with respect to both the baseline measure recorded in darkness and the pre-test. We will further investigate the determinants of the gradual build up and decay of the drift