Abstract
Disruption in sensory feedback can impair the brain’s ability to predict temporal and spatial congruence, often resulting in the perception of false stimuli. The sense of body ownership relies on the brain’s ability to integrate visual and tactile information across time through top-down processing and multisensory integration, allowing it to fill informational gaps. This study investigates how the synchronization of visual and tactile stimuli affects perceived ownership in the Rubber Hand Illusion (RHI) among neurotypical adolescents. Thirty right-handed students were randomly assigned to one of three conditions with varying levels of synchrony, and ownership was rated on a three-item Likert scale assessing responses to stylus, brush, and ice stimulation. Results showed that synchronized stimuli significantly enhanced the sense of ownership over the rubber hand, supporting theories of neuroplasticity and Bayesian sensory integration. By visualizing and sensing the same stimulus, participants perceived the rubber hand as their own hand. These findings provide baseline evidence on body ownership and temporal synchrony, suggesting approaches for future research that incorporate physiological measures.
Keywords: Rubber Hand Illusion, Neuroplasticity, Sensory Perception, Cognitive Integration, Predictive Coding, Multisensory Integration, Body Ownership, Temporal Synchrony
Introduction
Through traumatic injury, amputation, or congenital defects, over 35 million people worldwide (about twice the population of New York) use artificial limbs1. These prosthetics must be mentally and physically integrated into the body, but even a physically perfect device may not feel right. They become a part of their real body, pushing individuals to adjust to them despite the discomfort. Successful embodiment requires the development of a perceptual link between a biological and artificial limb. A functioning prosthetic uses the brain’s generated electrical signals which travel through nerves and are picked up by the prosthesis’ sensors. The motor cortex (located in the frontal lobe) sends signals to initiate movement and control various parts of the body, and the somatosensory cortex processes sensory information as described by Figure 1. These inputs run in the dorsal column–medial lemniscus pathway, synapsing in the thalamus before being projected to the primary somatosensory cortex (postcentral gyrus), where proprioceptive and tactile data are integrated2. The alignment of visual, tactile, and proprioceptive cues gives rise to the sense of body ownership, emerging from the integration of information across several sensory systems. When these factors do not align, the brain may produce false sensations or misattribute body parts. Experiments such as the Rubber Hand Illusion (RHI) produce evidence pointing to how temporal alignment can shape ownership3. This study explores this very phenomena, assessing the brain’s ability to be readily adaptive by using prior knowledge and experiences to predict incoming sensory information by the interpretation of sensory data through the use of the Rubber Hand Illusion.
Proprioception allows the brain to understand body movement through receptors in muscles and joints, sending signals which travel through fibers through the dorsal column pathways to the thalamus and somatosensory cortex. These signals are then integrated with visual and tactile inputs an individual experiences, maintaining a coherent sense of the body’s positions. In amputees, the disturbance of these signals triggers cortical reorganization in the somatosensory cortex, often resulting in altered proprioceptive feedback or phantom sensation. One of the most relevant theories based on the concept of ownership of a limb is the top-down processing model, describing how the brain makes use of already established information to construct a model reflecting the whole perceptual experience. This can be supported by the Bayesian Brain Theory which explains how the nervous system integrates information from different sensory outputs to fill in information and understand our environment. The theory supports the brain’s ability to compare expected sensory input with incoming data and update its body schema based on prior information. Figure 2 demonstrates how Bayesian integration adjusts its internal model to minimize prediction error, contributing to the sense of body ownership. The Bayesian Brain Theory elucidates this notion of the nervous system being able to operate in situations of uncertainty, such as being able to create false precepts in the absence of a familiar stimulus4. When visual and tactile stimuli are experienced in close temporal alignment, the brain can interpret the events as originating from the same source. Coherence, however, deteriorates in response to delayed or disconsonant stimuli, demonstrating the brain’s adaptive nature in relation to body representation.
Previous research on the Rubber Hand Illusion has demonstrated that ownership perception is strongly dependent on the temporal and spatial congruence of visuo-tactile signals. Asynchronous presentation of these signals weakens the illusion, highlighting the brain’s sensitivity to timing differences in multisensory integration. Though most studies have focused data on adult populations; the narrowed age group leaves less information about how the mechanism operates in adolescents- a group characterized by high neuroplasticity- and how body representation continues to adapt in developmental stages parallel to experience. Therefore, this experiment aims to evaluate how manipulation of synchronization levels influence the perceived sense of ownership in the RHI. By supporting the theory, synchrony was tested, and a connection was attempted to be established between the brain and rubber hand. By seeing a stimulus inflicted upon the rubber hand but feeling it on the real hand, the brain attempts to integrate this visual and tactile information, leading to the perception that the rubber hand is part of their body. This will be assessed by manipulating the synchronization by which the stimuli will be elicited. The synchronization of stimuli will vary by experiencing different timing conditions: synchronous (simultaneous stimuli), asynchronous 0.5 second delay, and asynchronous 3.0 second delay. This will ultimately be evaluated by a structured questionnaire to measure perceived ownership.
Variables
Independent variable: Temporal synchrony – time (seconds) between elicitation of visual and tactile stimulus on biological vs. rubber hand (0s, 0.5s, 3s).
Dependent variable: level of ownership felt towards the rubber hand (rated 1-7 on Likert Scale)
Experimental Hypothesis: If the placement of the stimuli is synchronized (inflicted at the same time – 0s delay), it will produce a stronger sense of ownership between the prosthetic and biological limb, describing how the brain’s capacity for neuroplasticity and multisensory integration, in which temporal alignment enhances information intake.
Null hypothesis: The synchronization of the stimuli will have no effect on the sense of ownership over the rubber hand.
Similar to the traditional design of the Rubber Hand Illusion, this experiment specifically utilizes the manipulation of synchronization to assess how correlated multisensory signals shape the feeling of ownership5. Through the time of the onset of the stimulus, the position of it seen and felt, illusory distortions can be experienced with the change of shape, size, and possibly color of the rubber hand to match a person’s true limb6. Based on the extent that the brain adjusts itself to the visual and tactile stimulation, the ventral premotor cortex anatomically connects with higher-order somatosensory areas post the receival of proprioceptive input. This process aligns with multi-sensory perception theory, which explains how the brain can integrate information from different sensory outputs (touch, sight, hearing, taste, smell) to fill in information and understand their environment7. This theory is consistent with the contribution of the ventral premotor cortex to self-attribution of body parts when being able to recognize and differentiate between limbs8. These mechanisms demonstrate the neural processes underlying body ownership and highlight the capacity of perceptual systems to adapt altered sensory environments. The results observed in this study may establish a foundation for implications that could benefit research and experimentation on prosthetic limbs, informing technological advancements that enhance embodiment and user experience. By applying the Rubber Hand Illusion concept to a group of neurotypical adolescents, this study investigates how multisensory integration and temporal synchrony affect perceived ownership.
Methods
This experiment was conducted with thirty right-handed 10th-grade students recruited through opportunity sampling from chemistry classes. This restricted the sample to neurotypical adolescents with similar cognitive abilities and age. Participants were randomly assigned into the three groups (synchronous, 0.5 second delay, or 3.0 second delay) to reduce expectancy bias through a self-programmed computer-based timer, which also standardized stimulus synchronization. To guarantee uniformity across all trials, the timer was programmed to indicate precise delay intervals between visual and tactile events. Although the tactile stimuli were manually applied by a researcher, the program’s automated system reduced variability. This system offered dependable temporal control within the setup’s constraints, despite achieving semi-automation rather than full.
The experiment follows an independent-groups design, ensuring the prevention of cross condition influence- responses from one condition would not influence another. Gender distribution was not controlled and is acknowledged as a limitation due to possible perceptual variation. The illusion was evaluated based on three different stimuli (tap, brush, temperature), all administered to the left hand of every individual. The full procedure was rehearsed prior to its conduction to ensure uniform timing, positioning, and application pressure throughout the experiment.
In order to keep the results as organic as possible, the designed ownership questionnaire avoids leading questions or suggestive wording and is answered from 1-7 based on the Likert scale for each item. The applied pressure of the stimuli was kept the same across participants, as variability may influence the strength of the illusion.
Although the results of the experiment will differ person to person and may be affected by the expectancy bias from participants wanting the experiment to work, the procedure is designed to minimize confounding variables. Individual differences including different levels of susceptibility to the illusion based on prior experience, psychological and cognitive differences will also be present, along with the degree to which the participants will give attention to the rubber hand. To further reduce expectancy bias, participants were provided with neutral instructions that emphasized sensory observation, without referencing the RHI phenomena or ownership.
All participants are given identical instructions and assent forms (Appendix iii, ii). The procedure of the experiment is clearly explained to eliminate deception and minimize confusion, and participants will be thoroughly debriefed when the study has been concluded (Appendix iv). The illusion was induced with three differing stimuli to generalize findings across sensory modalities and ensure effects are not specific to a type of catalyst. Using different stimuli strengthens the theory used for the experiment, allowing the determination of the brain’s processes, specifically of the integration of sensory information. This can allow the exploration of specific mechanisms in the illusion. Moreover, the variability and reliability of the experiment can be increased by the different materials to avoid confounding variables by ensuring that the effects observed by the illusion are from the intended manipulation rather than an inadvertent factor. During the debriefing stage, participants’ reports of proprioceptive drift- their perceived hand position in relation to the rubber hand- were qualitatively observed. Despite the lack of quantitative drift measurements, this observation subsidizes ownership responses obtained from the questionnaire.
The participant’s left hand will be placed in a cardboard box so it is not visible to them, and their right hand will be on a table so that it is parallel to a realistic (as possible) rubber hand. As shown in Figure 3, a blank cloth will cover the area between the prosthetic and the left shoulder to create a visual connection. Based on the assigned condition, a researcher will apply tactile stimuli (brush, tap, ice) to both the rubber hand and their real hand. Synchronous: simultaneous stimulus on both hands; asynchronous 0.5s: stimuli on biological hand followed by rubber hand after 0.5s; asynchronous 3.0s: stimuli on biological hand followed by rubber hand after 3.0s. All stimuli will be applied to corresponding locations on the participant’s real and rubber hands on the radial region of the dorsum (between the second and third metacarpal bones) to maintain spatial congruence. This will go on for a duration of 15 seconds with two-minute intervals between each stimulus, followed by the seven statements that will be read aloud for them to respond to from 1-7 based on how much it applies to their experience (strongly disagree → strongly agree) (Appendix v- full procedure):
| Control/Not control | Statement (Stylus tap) | Statement (Brush) | Statement (Ice) |
| Not control | I felt as if the prosthetic was my hand. | I felt as if the prosthetic was my hand. | I felt as if the prosthetic was my hand. |
| Not control | It seemed as if the sensation I felt was caused by the sensation I saw on the prosthetic hand. | It seemed as if the brush sensation I felt was caused by the sensation I saw on the prosthetic hand. | It seemed as if the sensation I felt was caused by the sensation I saw on the prosthetic hand. |
| Not control | It seemed like I was directly looking at my own hand rather than a prosthetic. | It seemed like I was directly looking at my own hand rather than a prosthetic. | It seemed like I was directly looking at my own hand rather than a prosthetic. |
| Not control | It seemed like the prosthetic hand was a part of my body. | It seemed like the prosthetic hand was a part of my body. | It seemed like the prosthetic hand was a part of my body. |
| Control (not considered for data) | It felt as if my right hand was drifting towards the prosthetic hand. | It felt as if my right hand was drifting towards the prosthetic hand. | It felt as if my right hand was drifting towards the prosthetic hand. |
| Control (not considered for data) | It appeared as if the prosthetic limb was drifting towards my right hand. | It appeared as if the prosthetic limb was drifting towards my right hand. | It appeared as if the prosthetic limb was drifting towards my right hand. |
| Control (not considered for data) | The prosthetic hand started to change shape, color, and appearance so that it started to resemble my right hand (visually) | The prosthetic hand started to change shape, color, and appearance so that it started to resemble my right hand (visually) | The prosthetic hand started to change shape, color, and appearance so that it started to resemble my right hand (visually) |
Analysis & Results
Out of the seven questionnaire statements, the first four were considered for our data to assess illusion strength, while the remaining three served as control questions as distractions to detect bias. These statements measured how strongly the prosthetic hand influenced the participant’s perception, which were ultimately averaged out per group to analyze the most successful group. The closer the value is to 7, the stronger the ownership towards the prosthetic hand. Internal consistency of the three-item ownership measure was excellent (Cronbach’s α = 0.95; item range = 0.89 – 0.94), indicating a strong reliability across stimulus type. Groups were labeled as Sync (synchronous), Async5 (0.5s delay), Async3 (3.0s delay).
| Synchronous | Asynchronous (0.5 sec.) | Asynchronous (3.0 sec.) | |
| Mean | 5.45 | 3.23 | 2.32 |
| Range | 4.34 | 4.5 | 2.42 |
| Standard Deviation | 1.40 | 1.44 | 0.95 |
The mean scores between Sync (5.45), Async5 (3.23), and Async3 (2.32) were distinctly different, suggesting a correlation between stimulus timing and perceived ownership. Group means (± SD, 95 % CI) were: synchronous 5.45 ± 1.40 [4.45 – 6.45]; Async5 3.23 ± 1.44 [2.20 – 4.26]; and Async3 2.32 ± 0.95 [1.64 – 3.00] (Appendix xi)9. Based on the data, participants in the Sync condition felt the greatest connection with the rubber hand, followed by Async5, then lastly Async3. This proves that shorter temporal delays between visual and tactile stimulus contribute to greater ownership perception. Additionally, although the Sync and Async5 groups had relatively similar standard deviations (1.40, 1.44 respectively), Async3 had a much lower standard deviation (0.95) which may imply that the data is comparatively less spread out. Because all three standard deviations were low, this suggests that the data can be considered reliable when interpreting it, although further inferential statistics were considered for validation of the results.
(Appendix vii: specific statistics of graph)
Prior to the conduction of inferential analysis, data was evaluated for assumptions through Shapiro-Wilk tests. This test confirmed that ownership ratings were normally distributed for each condition, with no strong departures from normality in the Async5 group (W = .97, p= .931) or Async3 (W = .88, p =. 139). Though the synchronous group showed a mild deviation (W = .82, p = .030), conduction of Levene’s test indicated homogeneity of variance across groups (F(2, 27) = 0.69, p = .51) (Appendix x). Given that the ANOVA is robust to mild normality violations, the analysis proceeded. The results of a conducted One-way ANOVA test showed that the p-value of the data was p = 2.37e-05, p < 0.05; F (2,27) = 16.21, p < 0.001, with a large effect size (partial η² = .55 [.24, .68], ω² = .50, Cohen’s f = 1.01) (Appendix xi)10. This shows that the data is statistically significant, meaning the null hypothesis can be rejected. A Tukey HSD Post-Hoc Testing was then conducted to reduce the risk of false positives and determine which groups differ, revealing significant differences between both Sync and Async5 (1.54e-03), as well as Sync and Async3 (1.01e-3), but not between Async5 and Async (Appendix ix).
These findings support that the hypothesis can be accepted, and the null hypothesis can simultaneously be rejected, indicating that placing the stimuli synchronously on the biological and rubber hands constructs the strongest sense of ownership. Thus, it can be said that synchronization significantly strengthens the perception of ownership of an artificial limb.
Conclusion and Methods
Sustaining sensory deafferentation can disrupt the brain’s ability to make predictions based on temporal and spatial incongruity. The reorganization of the sensory cortex which processes and makes sense of sensory information can compensate for missing information and activate areas of the brain crucial for stimulus detection. For the purpose of reducing uncertainty correlated to deafferentation, the sensory cortex’s hyperactivity will result in the perception of false information (e.g. feeling a sensation that is not actually on your own hand). This experiment and its results are consistent with the the top-down processing model and Bayesian integration by describing the brain’s capacity to use existing knowledge and multisensory information to fill in perceptual gaps. The current study examined perceptual aspects of visuo-tactile synchrony in the RHI, without testing sensorimotor feedback loops. Real prosthetic embodiment involved bidirectional integration of motor commands, and sensory inputs, which our design could not model11. The Rubber Hand Illusion demonstrates how the brain integrates visual and sensory perception into a cohesive representation when experienced simultaneously, dropping prediction error and enabling ownership of a rubber hand as part of one’s own body.
The highest mean (scale of 1-7) observed in the synchronous group (5.45) displays the link that can be developed when visualizing and sensing a stimulus at the same time. As the synchronization of the catalyst increases, the participant’s responses to the questionnaire approach 7, reflecting stronger ownership perception. The group with the most time between the stimuli being presented on the rubber and biological hands had the least average in the questionnaire of only 2.32, less than half of that of the synchronous condition. These findings can be explained through the top-down processing model akin to the Bayesian Brain Theory, supporting that the alignment of the sensory inputs’ coordination harmonized with the brain’s expectation for temporal congruence can strengthen perceptual binding12. The visual and tactile inputs allowed for the integration of the rubber hand into their established body schema, proposing the brain’s adaptation of its predictions based on preexisting ideas and new stimuli. As illustrated in Figure 4, synchronized stimuli produced stronger multisensory integration, resulting in an increased ownership perception. This supports the hypothesis that synchronized conditions induce greater ownership towards the rubber hand as the brain adjusts its standards to incorporate it into its internal model, ultimately creating a robust perceptual shift. The difference in time between the sensations in both asynchronous groups establish less consistent sensory information, weakening integration and embodiment.
This study’s independent measures allow the elimination of order effects since each participant was assigned to a single condition. Because of this, participants could not infer the hypothesis, reducing the possibility of demand characteristics or a placebo effect. The mix of gender and backgrounds of participants broadens the generalizability of the study’s and improves population validity. Another strength of the design is the consistent procedure (Appendix v) and standardized instructions to all participants. No participants were told anything different to reduce variability in the interpretation of the experiment, increasing internal validity. One key strength of the design was the use of three different stimuli which enhanced the reliability of the data by making certain that the illusions’ observed effects resulted from the deliberate manipulation, rather than unintended variables. Lastly, the inclusion of three control questions in the questionnaire not considered for the data minimized response bias by discouraging participants from guessing the purpose of the experiment and adjusting their answers to align with the optimal response.
Although the findings reflected high internal consistency, generalizability can be limited by a number of factors. The opportunity sample of tenth-grade students diminishes its external validity, as age, developmental stage, and cultural background can influence perceptual integration. Adolescent neuroplasticity may introduce variability in the perception of ownership. The familiar school setting where the experiment took place may deem ecological validity low, since it does not reflect clinical or everyday sensory environments. Furthermore, given known gender-related differences in multisensory integration, the skewed gender distribution among the subjects possibly impacted perceptual responses. Population validity and possibility for the results to be extrapolated to a larger population may be attained with a more representative and gender-balanced sample.
The results are based on perceptual mechanisms rather than true clinical embodiment, as the study involved healthy individuals using a simulated. They may not be the same for people who have limb loss and have changed sensory feedback. Additionally, because the experiment took place in a single session, no conclusions can be made concerning adaptation or neuroplasticity. In future studies, multi-session protocols should be used in order to examine the effects of repeated exposure on the strength of ownership and integration at the neural level. Visual mismatches, for example, the participant’s skin tone differing from the rubber hand’s, may possibly lead to a different perception, aligning with prior evidence that visual congruence improves embodiment.
Moreover, the measure of ownership exclusively utilized scores in self-ratings, which are subject to expectancy or response bias. Proprioceptive drift and subjective ownership are known to sometimes dissociate, and including physiological measures such as skin conductance, an EEG, or EMG could provide more insight into the neural processes underlying the illusion13. Although drift measurements have technical limitations, combining them with subjective ratings and physiological measures can provide a more comprehensive assessment of body ownership14.
Additionally, the extent of a participant’s attention towards the experiment is a limitation, because a person concentrating on sensory experience more may feel a greater ownership towards the prosthetic than those who were less attentive. This variability may have influenced responses to the questionnaire, potentially reducing accuracy. Moreover, although stimulus synchrony was controlled by a semi-automated computer program, subtle inconsistencies in timing could still influence the results. As a modification, future experimentation could implement a full mechanical system for maximum accuracy in removing potential confounding variables. Also, the study’s limited delay conditions do not capture the whole temporal binding window for visuospatial integration fully. Adding more specific delay intervals and adding negative controls like spatial mismatches, visual-only stimuli, or longer asynchronous delays, can allow for clearer distinguishment of more temporal and synchrony-specific effects from more general multisensory stimulation.
Ultimately, optimizing synchronization between sensory inputs advances more robust multisensory integration and coherence for the body-schema, illustrating the way in which timing and perception in combination defines the brain’s sense of embodiment. Therefore, while the current work provides insights into temporal synchrony and body representation, the inferences should be limited to models of ownership at the perceptual level.
Discussions
Modern day technology offers amputees a wide variety of choices in function and purpose. Body powered prostheses are controlled by the body, and these can be especially useful for a hand amputation to restore the ability to pick up objects and even engage in manual labor15. Second, a motor-powered prosthesis has buttons that controls movements, such as one to articulate finger movements to pick up an object or point to something. Lastly, the recent development of myoelectric powered prosthetics, the most advanced of the three, allows controlling prosthetic limbs by electric signals sent to the device through electrodes placed on the skin16. The degree of sensory feedback these systems offer varies, with myoelectric most closely resembling a biological limb in terms of its timing and responsiveness. The synchrony principle tested in this experiment, where simultaneous visual and tactile cues improved ownership perception, is mirrored in this temporal precision.
While the present study did not involve prosthetic use, its findings provide perceptual evidence relevant to future prosthetic integration. The demonstrated link between visual and tactile stimulation suggests that improving embodiment of artificial limbs depends on enhancing temporal congruence between multisensory feedback signals. Further research integrating real-time sensory in myoelectric or hybrid systems could help bridge the perceptual gap between artificial and biological movement, potentially allowing simultaneous movement across multiple limbs. The findings of this study suggest optimizing between sensory inputs can especially enhance the brain’s connection with an artificial limb, supporting the development of prosthetics that better replicate natural perception and control.
Appendix i: Parental Consent Form
Appendix ii: Assent Form
Appendix iii: Standardized Instructions
Appendix iv: Standardized Debriefing Notes
Appendix v: Procedure
Appendix vi: Raw Data
Synchronous condition
Asynchronous (0.5 seconds) condition
Asynchronous (3.0 seconds) condition
Appendix vii: Statement Responses (Raw Data)
Synchronous condition
Asynchronous (0.5 seconds) condition
Asynchronous (3.0 seconds) Condition
Appendix viii: Box Plot Statistics
| Sync | Async5 | Async3 | |
| lower quartile (Q1) | 5.33 | 2.08 | 1.33 |
| upper quartile (Q3) | 6.42 | 4.08 | 3.33 |
| IQR | 1.09 | 2 | 2 |
| median | 5.8 | 3.33 | 2.13 |
| outliers | 2.33 | none | none |
Appendix ix: One-way ANOVA test and Tukey HSD Test
Appendix x: Shapiro-Wilk Test and Levene’s Test
Synchronous condition
Asynchronous (0.5 seconds) condition
Asynchronous (3.0 seconds) condition
Levene’s Test
Appendix xi: Confidence Interval and Effect Size Calculations
Synchronous condition
Asynchronous (0.5 seconds) condition
Asynchronous (3.0 seconds) condition
Effect size
References
- World Health Organization. Standards for prosthetics and orthotics. World Health Organization (2017). [↩]
- D. Purves, G. J. Augustine, D. Fitzpatrick, et al. Neuroscience. 2nd ed. Sunderland (MA): Sinauer Associates; The major afferent pathway for mechanosensory information: The dorsal column–medial lemniscus system. https://www.ncbi.nlm.nih.gov/books/NBK11142/ (2001). [↩]
- M. Botvinick, J. Cohen. Rubber hands ‘feel’ touch that eyes see. Nature. 391, 756 (1998a). [↩]
- H. Bottemanne. Bayesian brain theory: Computational neuroscience of belief. Neuroscience, 566, 198-204. (2025). [↩]
- J. Cohen, M. Botvinick. Rubber hands ‘feel’ touch that eyes see. Nature. 391, 756. (1998b). [↩]
- J. Zbinden, M. Ortiz-Catalan. The rubber hand illusion is a fallible method to study ownership of prosthetic limbs. Scientific Reports. 11(1). (2021). [↩]
- H. H. Ehrsson, N. P. Holmes, R.E. Passingham. Touching a rubber hand: Feeling of body ownership is associated with activity in multisensory brain areas. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience. 25(45), 10564–10573 (2005). [↩]
- M. S. Graziano, C. G. Gross. Spatial maps for the control of movement. Current Opinion in Neurobiology. 8(2), 195-201 (1998). [↩]
- StatsKingdom. ANOVA and effect size calculators. https://www.statskingdom.com/ (2024). [↩]
- J. O. Uanhoro, J. O. Effect size calculators. https://effect-size-calculator.herokuapp.com/ (2017). [↩]
- T. R. Clites, M. J. Carty, J. B. Ullauri, M. E. Carney, L. M. Mooney, J. F. Duval, S. S. Srinivasan, H. M. Herr. Proprioception from a neurally controlled lower-extremity prosthesis. Science Translational Medicine. 10(443), eaap8373 (2018). [↩]
- C. D. Gilbert, W. Li. Top-down influences on visual processing. Nature Reviews. Neuroscience. 14(5), 350–363. (2013). [↩]
- G. Tosi, B. Mentesana, D. Romano. The correlation between proprioceptive drift and subjective embodiment during the rubber hand illusion: A meta-analytic approach. Quarterly Journal of Experimental Psychology. 76(10), 2197–2207 (2023). [↩]
- W. Zhang, Y. Shi, Y. Peng, L. Zhong, S. Zhu, W. Zhang, S. J. Tang. Neuron activity-induced Wnt signaling up-regulates expression of brain-derived neurotrophic factor in the pain neural circuit. The Journal of Biological Chemistry. 293(40), 15641–15651 (2018). [↩]
- Arm Dynamics. Introduction to Body-Powered Prostheses. https://www.armdynamics.com/upper-limb-library/introduction-to-body-powered-prostheses (2021). [↩]
- A. Henson. Introduction to Myoelectric Prostheses. Arm Dynamics. https://www.armdynamics.com/upper-limb-library/introduction-to-myoelectric-prostheses [↩]
