Date Published: February 28, 2019
Publisher: Public Library of Science
Author(s): Arvid Guterstam, Dennis E. O. Larsson, Hugo Zeberg, H. Henrik Ehrsson, Matthew Longo.
Can the mere expectation of a sensory event being about to occur on an artificial limb be sufficient to elicit an illusory sense of ownership over said limb? This issue is currently under debate and studies using two different paradigms have presented conflicting results. Here, we employed the two relevant paradigms, namely, the magnetic touch illusion and the “tactile expectation” version of the rubber hand illusion, to clarify the role of tactile expectations in the process of attributing ownership to limbs. The illusory senses of ownership and ‘magnetic touch’ were quantified using questionnaires, threat-evoked skin conductance responses and a combination of motion tracking synchronized with real-time subjective ratings and skin conductance. The results showed that the magnetic touch illusion was dependent on concurrent visual and tactile stimulation and that visually induced tactile expectations alone were insufficient. Moreover, in this study, tactile expectations were not associated with the rubber hand illusion, neither in terms of subjective ratings nor skin conductance changes. Together, these findings contradict the notion that the brain uses predictions of upcoming sensory events to determine whether or not a limb belongs to the self, and, instead, emphasize the importance of correlated multisensory information.
When reaching out your hand to catch a ball thrown at you, the brain forms predictions about the upcoming tactile event based on the visual input from the incoming ball and your own hand. In certain situations, sensory predictions may even attenuate the perception of the subsequent tactile event. For instance, it has been shown that a self-generated tactile stimulus is perceived as weaker [1,2] and less ticklish  than the same stimulus generated externally. In the body ownership literature, there is currently an ongoing debate relating to whether sensory predictions may alter the perception of an artificial limb. Specifically, it has been argued that the mere expectation of a tactile event being about to occur on a rubber hand is sufficient to elicit a sense of ownership over that limb [4–6], which contradicts the leading view in the field that ownership illusions require correlated sensory signals from at least two sensory modalities (or somatosensory submodalities) [7–11]. However, a recent study from our group investigating the “magnetic touch illusion” failed to find an effect of such “tactile expectations” on limb ownership , sparking a scientific debate . In this study, we set out to test whether tactile expectations play any role in the generation of magnetic touch illusion [12,14] as well as replicating the basic tactile expectation effect on ownership [4–6].
To examine the potential role of tactile expectations in magnetic touch illusion, we first analyzed the average ratings on the magnetic touch (S1-S2) and ownership questionnaire statements (S5-S6). The visuo-tactile condition was associated with significantly higher ratings of magnetic touch and ownership sensations compared to the visual only—brush approaching, visual only—hand approaching and visual only—hand static conditions respectively (magnetic touch statements: t23 = 4.40, p<0.001, d = 1.12; t23 = 3.04, p = 0.006, d = 0.72; t23 = 3.71, p = 0.001, d = 0.97; and ownership statements: t23 = 3.03, p = 0.006, d = 0.67; t23 = 2.76, p = 0.011, d = 0.69; t23 = 2.29, p = 0.031, d = 0.66; two-tailed paired t-tests; Fig 2A). These results suggest that the illusion was successfully induced in the visuo-tactile condition and that tactile expectations alone was not sufficient. As shown in Fig 2B and 2C, this subjective difference was mirrored in greater threat-evoked SCRs in the visuo-tactile compared to the hand approaching (mean difference±SEM: 0.039±0.015 μS; t23 = 2.54, p = 0.018, d = 0.52), hand static (mean difference±SEM: 0.041±0.019 μS; t23 = 2.13, p = 0.044, d = 0.44) and visual only conditions (mean difference±SEM: 0.029±0.016 μS; t23 = 1.85, p = 0.078, d = 0.38). We used the magnetic touch illusion and a combination of motion tracking, real-time subjective ratings and skin conductance responses to investigate the role of tactile expectations in the process attributing ownership to limbs. In summary, the results showed that the illusion is dependent on concurrent visual and tactile stimulation and that tactile expectations alone are not sufficient to induce either illusory ownership or a sense of magnetic touch, even when the trajectory of the visual stimulus is along a path compatible with a potential collision of both the real and rubber hands . Furthermore, we found no evidence, neither in terms of subjective ratings nor skin conductance responses, for the notion that tactile expectations can induce ownership of an artificial limb, even when using the same key elements of the experimental setup of a previous study that reported the existence of such an effect . These findings emphasize the importance of multisensory correlations for the emergence of limb ownership and contradict the hypothesis that sensory predictions of upcoming tactile events are sufficient to determine whether or not a limb belongs to the self. Source: http://doi.org/10.1371/journal.pone.0213265