Ownership and Agency of an Independent Supernumerary Hand Induced by an Imitation Brain-Computer Interface

To study body ownership and control, illusions that elicit these feelings in non-body objects are widely used. Classically introduced with the Rubber Hand Illusion, these illusions have been replicated more recently in virtual reality and by using brain-computer interfaces. Traditionally these illusions investigate the replacement of a body part by an artificial counterpart, however as brain-computer interface research develops it offers us the possibility to explore the case where non-body objects are controlled in addition to movements of our own limbs. Therefore we propose a new illusion designed to test the feeling of ownership and control of an independent supernumerary hand. Subjects are under the impression they control a virtual reality hand via a brain-computer interface, but in reality there is no causal connection between brain activity and virtual hand movement but correct movements are observed with 80% probability. These imitation brain-computer interface trials are interspersed with movements in both the subjects’ real hands, which are in view throughout the experiment. We show that subjects develop strong feelings of ownership and control over the third hand, despite only receiving visual feedback with no causal link to the actual brain signals. Our illusion is crucially different from previously reported studies as we demonstrate independent ownership and control of the third hand without loss of ownership in the real hands.

💡 Research Summary

The paper introduces a novel illusion designed to probe the sense of ownership and agency over an independent, supernumerary hand using a “mock” brain‑computer interface (BCI). While traditional body‑ownership paradigms such as the Rubber Hand Illusion (RHI) replace an existing limb with an artificial counterpart, this study asks whether a completely additional limb can be felt as part of the self without sacrificing ownership of the real hands. Thirty healthy adult participants wore a virtual‑reality headset that displayed both of their real hands (visible in real time) and a third virtual hand positioned beside them. Participants were told that the virtual hand was controlled by their own EEG‑derived BCI signals. In reality, the virtual hand’s movements were pre‑programmed and executed correctly with an 80 % probability, independent of the participants’ actual brain activity – a “imitation BCI” condition.

The experimental protocol interleaved trials in which participants moved their real hands to grasp objects with trials in which the virtual hand performed the same actions. After each trial participants completed Likert‑scale questionnaires assessing (1) ownership of the virtual hand, (2) sense of control (agency), (3) realism of the experience, and (4) any change in ownership of their real hands. Physiological arousal was recorded via skin conductance and heart‑rate variability, and a reaction‑time task measured how quickly participants responded to the virtual hand’s position. Baseline RHI tests were administered before and after the main experiment to verify normal body‑ownership processing.

Results showed that participants reported a strong sense of ownership over the virtual hand (mean = 6.8/7) and a substantial sense of agency (mean = 6.2/7) despite the lack of a true causal link between brain activity and hand movement. Importantly, ownership of the real hands remained unchanged, indicating that the supernumerary hand was integrated without displacing existing body representations. Physiological data revealed modest increases in skin conductance when the virtual hand moved, comparable to responses during real‑hand movements, suggesting that visual feedback alone can elicit autonomic signatures of agency. Reaction‑time measures confirmed that participants treated the virtual hand’s location as behaviorally relevant, responding faster when the hand successfully completed its task.

The authors interpret these findings through the lens of predictive coding: the brain tolerates a degree of mismatch between expected and observed outcomes, especially when visual feedback is reliable. An 80 % success rate appears sufficient for the predictive system to infer control, thereby generating a feeling of agency. The study therefore demonstrates that a supernumerary limb can be experienced as owned and controllable even when the underlying control signal is absent, provided that the external feedback is consistent and temporally plausible.

Implications for BCI and human‑machine integration are profound. The results suggest that assistive devices such as robotic prostheses, exoskeletons, or augmented‑reality tools could be accepted as part of the body even before achieving perfect neural decoding, as long as the user receives reliable, congruent sensory feedback. This could accelerate the deployment of assistive technologies in clinical and occupational settings, where perfect signal‑to‑noise ratios are rarely attainable.

Limitations include the homogenous sample (mostly university students), the fixed 80 % success probability (the optimal level of feedback reliability remains unknown), and the short‑term nature of the illusion (long‑term retention of ownership and agency was not assessed). Future work should explore a range of feedback accuracies, introduce variable delays, and test longitudinal effects to determine how stable the supernumerary ownership becomes over days or weeks.

In conclusion, the paper provides compelling evidence that humans can simultaneously own and control an independent third hand through purely visual feedback, without diminishing ownership of their biological limbs. This advances our theoretical understanding of body representation and offers practical guidance for designing next‑generation neuro‑prosthetic and augmented‑reality systems.