No causal link between changes in hand position sense and feeling of limb ownership in the rubber hand illusion

Abstract

The rubber hand illusion is a perceptual illusion in which participants experience an inanimate rubber hand as belonging to their own body. The illusion is elicited by synchronously stroking the rubber hand and the participant's real hand, which is hidden from sight. The feeling of owning the rubber hand is accompanied by changes in hand position sense (proprioception), so that when participants are asked to indicate the location of their (unseen) hand, they indicate that it is located closer to the rubber hand. This "proprioceptive drift" is the most widely used objective measure of the rubber hand illusion, and from a theoretical perspective, it suggests a close link between proprioception and the feeling of body ownership. However, the critical question of whether a causal relationship exists between changes in hand position sense and changes in limb ownership is unknown. Here we addressed this question by devising a novel setup that allowed us to mechanically manipulate the position of the participant's hand without the participant noticing, while the rubber hand illusion was being elicited. Our results showed that changing the sensed position closer to or farther away from the rubber hand did not change the strength of the rubber hand illusion. Thus, the illusion is not dependent on changes in hand position sense. This finding supports models of body ownership and central body representation that hold that proprioceptive drift and the subjective illusion are related to different central processes.

Keywords: Modularity of perception; Multisensory processing; Spatial localization.

Fig. 1

Schematic illustration of the setup employed in both experiments. The rubber hand was placed on top of a small table, and the participant’s hand was placed 12.5 cm underneath the rubber hand on the mechanical hand displacement apparatus. Point A indicates the position of the rubber hand, B and D indicate the different start positions of the participant’s hand, and C indicates the end position. The arrows indicate the positions between which the participant’s hand was displaced in the various conditions of the two experiments. In the static condition of Experiment 2, the participant’s hand remained static at Position B

Fig. 2

Ratings for the questionnaire statements on a 7-point Likert scale ranging from −3 to +3, with −3 corresponding to fully disagree and +3 corresponding to fully agree. The bars show the means for the pooled ratings from all participants for each statement and condition. The error bars indicate the standard errors of the means

Fig. 3

Results from the questionnaire (i.e., illusion index; A) and proprioceptive drift (B) for Experiment 1. The illusion index was calculated by subtracting the means of the ratings for the pooled control statements from the means of the ratings for the pooled illusion statements. Proprioceptive drift was measured in centimeters; values >0 were directed toward the rubber hand, and values <0 were directed away from the rubber hand. Error bars indicate the standard errors of the means. ^* p < .05, ^*** p < .001

Fig. 4

A descriptive illustration of all pointing responses in the proprioceptive drift test in Experiment 1. The indicator scale is in centimeters. Stars indicate the mean estimated right-hand positions, and the lines indicate the standard errors of the means. Both the responses prior to the period of visuotactile stimulation and mechanical hand displacement (“pre”; hollow stars) and the responses after this period (“post”; black stars) are indicated. (The proprioceptive drift corresponds to the difference between “post” and “pre,” as we describe in the Method section.) The 8-cm position indicates the starting position for the participant’s hand in the “away” conditions, whereas the −8-cm position indicates the starting position for the hand in the “toward” conditions. The 0-cm position indicates the final position of the hand in all conditions. The rubber hand was placed at 16 cm (not shown in the figure)

Fig. 5

Ratings for the questionnaire statements on a 7-point Likert scale ranging from −3 to +3, with −3 corresponding to fully disagree and +3 corresponding to fully agree. The bars show the means for the pooled ratings from all participants for each statement and condition. The error bars indicate the standard errors of the means

Fig. 6

Results from the questionnaire (i.e., illusion index; A) and proprioceptive drift (B) for Experiment 2. The illusion index was calculated by subtracting the means of the ratings for the pooled control statements from the means of the ratings for the pooled illusion statements. Proprioceptive drift was measured in centimeters; values >0 were toward the rubber hand, and values <0 were away from the rubber hand. Error bars indicate the standard errors of the means. ^* p < .05, ^** p < .01, ^*** p < .001

Fig. 7

A descriptive illustration of all pointing responses in the proprioceptive drift test in Experiment 2. The indicator scale is in centimeters. Stars indicate the mean estimated right-hand positions, and the lines indicate the standard errors of the means. Both the responses prior to the period of visuotactile stimulation and mechanical hand displacement (“pre”; hollow stars) and the responses after this period (“post”; black stars) are indicated. (The proprioceptive drift corresponds to the difference between “post” and “pre,” as we describe in the Method section.) The 8-cm position indicates the starting position for the participant’s hand in all conditions, whereas the 0-cm position indicates the final position in the “away-from” conditions. In the static conditions, the participant’s hand remained at 8 cm throughout the experiment. The rubber hand was placed at 16 cm (not shown)

Fig. 8

Correlations between illusion strength and proprioceptive shift in Experiment 2. Illusion strength was calculated by subtracting the means of the pooled ratings of the illusion statements in the asynchronous static condition from the means of the pooled ratings of the illusion statements in the synchronous static condition. Proprioceptive shift is indicated in centimeters

Fig. 9

Postexperiment questionnaire results. (A) Responses to the forced choice question Q3 (see Table 2). (B) Participants’ responses to Q4. The responses to Q4 are translated from the original Swedish into English