Brain Computer Interface Using Modulation of Auditory Steady-State Response with Help of Stochastic Resonance^.

Area of Science:

• Neuroscience and Biomedical Engineering
• Brain-Computer Interfaces (BCI)
• Auditory Neuroscience

Background:

• Traditional brain-computer interfaces (BCIs) often rely on eye movements, limiting their applicability, especially for individuals with severe motor impairments.
• The auditory steady-state response (ASSR) offers a potential modality for BCIs due to its robustness and distinct neural signatures.
• Developing eye-movement-independent BCIs is crucial for expanding assistive technology options.

Purpose of the Study:

• To propose and evaluate an eye-movement-independent BCI system utilizing the auditory steady-state response (ASSR-BCI).
• To investigate the efficacy of exploiting stochastic resonance to enhance signal separation within the ASSR-BCI.
• To determine the feasibility of a practical BCI for users in an eyes-closed state.

Main Methods:

• The proposed ASSR-BCI system uses amplitude-modulated (AM) tones, requiring users to selectively attend to one of two presented AM tones.
• A stochastic resonance effect was incorporated to improve the separation of neural signals elicited by the attended AM tones.
• Performance was evaluated across nine subjects under noisy conditions (60 dB tones, 30 dB noise) measuring classification accuracy and information transfer rate.

Main Results:

• The ASSR-BCI achieved a mean classification accuracy of 77% across nine subjects under the specified noisy conditions.
• Analysis of information transfer rate suggests that an inter-trial interval of 2-3 seconds is adequate for a given trial time length.
• The system demonstrated robustness in distinguishing attended auditory stimuli.

Conclusions:

• The developed auditory steady-state response brain-computer interface (ASSR-BCI) is effective and independent of eye movements.
• Exploiting stochastic resonance enhances the performance of the ASSR-BCI, making it more reliable in the presence of noise.
• The findings indicate the feasibility of developing a practical, eyes-closed BCI by optimizing parameters like trial time and electrode placement for individual users.