Fig 1.
A) The top panel shows the setup for our third hand illusion. The bottom panel shows our set up for the rubber hand illusion. B) The virtual third hand as seen by subjects during the third hand illusion. The leftmost and the middle panel show the starting and end position of the hand movement. The rightmost panel shows the ‘broken’ fingers at the threat. C) Block diagram of the experiment
Fig 2.
The questions and responses for the different experimental conditions.
Subjects were asked to complete the questionnaire immediately after each experimental condition using a seven-step visual-analogue scale ranging from strongly agree (+++) to strongly disagree (---). The black dots indicate the median responses and the colour scale indicates the percentage of the total responses. For the real questionnaires given to subjects the [condition] of each statement was substituted with “virtual” for IBCI and “rubber” for RHI or RTHI. See also Table 1 for statistical comparisons between different experimental conditions.
Fig 3.
The questions and responses for the different baseline conditions.
Subjects were asked to complete the questionnaire immediately after each baseline condition using a seven-step visual-analogue scale ranging from strongly agree (+++) to strongly disagree (---). The black dots indicate the median responses and the colour scale indicates the percentage of the total responses. For the real questionnaires given to subjects the [condition] of each statement was substituted with “virtual” for IBCI and “rubber” for RHI or RTHI. See also Table 2 for comparisons between baseline and illusion conditions.
Table 1.
Statistical assessment of differences between conditions for each question.
Table 2.
Statistical comparison with baseline conditions.
Fig 4.
A) The mean across subjects GSR after a threatening stimulus during the IBCI condition. A significantly stronger response was seen during the imitation of BCI control compared to the pre and post conditions where subjects knew there was no control. Error bars indicate the standard error of the mean (SEM). B) Correlation between subject’s responses to Q3 (addressing feeling of control) and Q1 and Q2 (both addressing the feeling of ownership) during the IBCI condition. As subjects answers for Q1 and Q2 were highly correlated within the IBCI condition (R = 0.87, p<0.001, Spearman’s rank correlation) the data from both questions has been pooled. Hence, for each subject the plot contains one paired sample representing the subject’s answers to Q3 and Q1 and a second paired sample representing the answers to Q3 and Q2. The colour scale indicates the number of responses. C) Mean across subjects difference in temperature between the subjects real hands aligned to illusion onset (see Materials and Methods for details). Shaded areas show the SEMs.