Influence of interoception and body movement on the rubber hand illusion

Abstract

Rubber hand illusion (RHI) refers to the illusory sense of body ownership of a fake hand, which is induced by synchronous visuotactile stimulation to the real and fake hands. A negative correlation was reported between the cardiac interoception and the strength of RHI, but the subsequent studies have been unsuccessful in replicating it. On the other hand, voluntary action is suggested to link interoception and the sense of body ownership in different situations. If so, moving RHI, induced by the active or the passive finger tapping while observing a fake hand, might reveal the relationship more clearly. The measurement of interoception has been another issue. We, therefore, examined the relationship between the moving RHI and two measures of interoception: interoceptive accuracy (IAcc) measured by the conventional heartbeat counting task and interoceptive sensibility (IS) measured using a questionnaire. For the classical visuotactile RHI, our results supported the lack of association between the interoception measures and RHI. For the moving RHI, a stronger sense of body ownership was induced for participants with higher IS regardless of active or passive movement, and a stronger sense of agency was caused by active than passive movement only for those with lower IAcc. These results reveal the dynamic links between the interoception and the bodily senses. The results also suggest that multiple dimensions of interoception affect the bodily senses differently.

Keywords: body movement; interoception; rubber hand illusion; sense of agency; sense of body ownership.

Figure 1

The setup and conditions of RHI experiments. (A) The fake hand was placed on top of the box, while the participant’s hand was placed inside and below. (B) The paper clipboard for measuring the proprioceptive drift (removed in the other pictures for clarity.) (C) The classical RHI stimulation with a paintbrush. (D) The moving RHI stimulation. The connecting stick was adjusted in each experimental block so that the participant or the experimenter moved the fake hand’s finger to be synchronous or asynchronous with the participant’s finger. A summary of trial conditions is shown to the right of parts (C) and (D).

Figure 2

(A) The results of interoceptive measures of IS and IAcc are shown as a scatter plot. Each marker represents a single participant. The dashed line shows the linear regression line, with the gray shades showing the 95% confidence intervals. The outlined symbols represent the outliers that were to be excluded in the later analyses. (B) The plot of rating scores under the three RHI conditions is plotted separately for synchronous and asynchronous stimulation. The error bars show standard errors of the mean across participants. SBO, sense of ownership, SA, sense of agency, ctl, control. *p < 0.05; **p < 0.01; ***p < 0.001. (C) A scatter plot of SBO and SA. Each marker represents a participant, but a participant appears twice for the active and passive conditions of moving RHI. The outlined symbols show the outliers that were excluded in the analyses. The dashed line shows the linear regression line (without the outliers) with the gray shades showing the 95% confidence intervals. (D) A plot of proprioceptive drift (PD) for synchronous and asynchronous stimulations for each RHI condition.

Figure 3

(A) The plot of the SBO index for high and low IAcc or IS for each RHI condition. (B) The plot of the SA index is plotted in the same way as the SBO index. Two outliers (see text) were excluded in these plots. The error bars show standard errors of the mean across participants. +, p < 0.1; *p < 0.05; **p < 0.01.