Introduction
This document serves as a formal Author Correction for the research article titled "Neural signatures of motor imagery for a supernumerary thumb in VR: an EEG analysis," which was originally published in Scientific Reports in September 2024. The primary objective of this update is to rectify an inaccuracy in the "Data availability" section, ensuring that the global research community has transparent and immediate access to the underlying datasets.
Research Context: Human Augmentation and BCIs
The original study addresses a burgeoning field in neuroengineering: human augmentation. Specifically, it investigates the brain's ability to generate distinct neural signatures for the control of a "supernumerary" or extra limb. In this experimental setup, participants utilized a virtual reality (VR) environment to perform motor imagery—the mental representation of movement—for a virtual third thumb.
Understanding how the primary motor cortex and associated regions adapt to an additional appendage is critical for the development of advanced brain-computer interfaces (BCIs) and neuroprosthetics. If the brain can successfully differentiate the imagery of an extra limb from existing digits, it opens the door to seamless integration of robotic enhancements.
Methodology and Data Processing
While the core findings are detailed in the original 2024 publication, this correction highlights the specific nature of the data involved in the study's classification models:
- Modality: The researchers employed Electroencephalography (EEG) to capture high-temporal-resolution brain activity during the imagery tasks.
- Task Design: Participants engaged in motor imagery, imagining the movement of the virtual thumb without physical execution, which is a standard protocol for non-invasive BCI training.
- Signal Processing: The study utilized epoched EEG data. This involves segmenting continuous brain signals into discrete time windows (epochs) relative to the onset of the imagery task. These epochs serve as the primary input for machine learning classifiers designed to recognize the neural patterns of the supernumerary limb.
Correction to Data Availability
The original publication contained a statement indicating that the research data was under an embargo. This correction notice officially lifts that restriction. The authors have clarified that the data is now publicly available to facilitate reproducibility and further meta-analysis by other researchers in the field.
Dataset Details
The dataset has been hosted on the Open Science Framework (OSF) to ensure long-term accessibility and version control.
- Content: Epoched EEG data used for the study's classification models.
- Access Link: https://osf.io/qvbfg/
- Status: Publicly accessible; no embargo.
Conclusion
This correction underscores the commitment of the authors and Scientific Reports to the principles of open science. By providing the epoched EEG data, the researchers enable the broader scientific community to verify their classification results and explore new algorithmic approaches to the challenge of supernumerary limb control. This transparency is essential for advancing the state of the art in human-machine interaction and neural plasticity research.
Summary updated