NeuroPrompt-BCI: Selective Prompt Insertion and Word-Level Error-Aware Language Modeling for Adaptive Brain–Computer Interface Communication Systems
Keywords:
brain–computer interface, prompt tuning, language modeling, word-level error correction, adaptive systems, neural decodingAbstract
Brain–computer interface (BCI) systems enable individuals with severe motor disabilities to communicate by translating neural signals into textual output. Despite significant advances in neural decoding, these systems remain plagued by high error rates at the character and word levels, which degrade communication speed and user trust. Language models can provide powerful error correction, but their monolithic application often introduces latency and fails to account for user-specific and context-dependent error distributions. In this paper we propose NeuroPrompt-BCI, a novel system architecture that combines selective prompt insertion with word-level error-aware language modeling for adaptive BCI communication. Drawing on recent advances in prompt tuning, the system dynamically inserts lightweight, learnable prompt tokens into a pretrained language model only when decoding uncertainty or historical error patterns suggest a high risk of incorrect output. The prompt insertion policy is guided by a word-level error analyzer that models substitution, deletion, and insertion errors typical of specific BCI paradigms, learning user-specific confusion patterns over time. We discuss the full system stack, from neural signal acquisition and feature extraction through adaptive language generation, and analyze structural trade-offs among latency, computational overhead, accuracy, and personalization. We further address infrastructure deployment at the edge-cloud boundary, data governance for neural signals, fairness across diverse user populations, and robustness to signal non-stationarity. By integrating selective prompt insertion with fine-grained error awareness, NeuroPrompt-BCI offers a principled path toward more reliable, transparent, and user-sensitive BCI communication systems.
References
1. Farwell, L. A., & Donchin, E. (1988). Talking off the top of your head: Toward a mental prosthesis utilizing event-related brain potentials. Electroencephalography and Clinical Neurophysiology, 70(6), 510–523.
2. Speier, W., Arnold, C., & Pouratian, N. (2016). Evaluating true BCI communication rate through mutual information and language models. Journal of Neural Engineering, 13(4), 046019.
3. Willett, F. R., Avansino, D. T., Hochberg, L. R., Henderson, J. M., & Shenoy, K. V. (2021). High-performance brain-to-text communication via handwriting. Nature, 593(7858), 249–254.
4. Orhan, U., Hild, K. E., Erdogmus, D., Roark, B., Oken, B., & Fried-Oken, M. (2012). RSVP keyboard: An EEG based typing interface. In 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 645–648).
5. Makin, J. G., Moses, D. A., & Chang, E. F. (2020). Machine translation of cortical activity to text with an encoder–decoder framework. Nature Neuroscience, 23(4), 575–582.
6. Li, X. L., & Liang, P. (2021). Prefix-tuning: Optimizing continuous prompts for generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (pp. 4582–4597).
7. Liu, X., Zheng, Y., Du, Z., Ding, M., Qian, Y., Yang, Z., & Tang, J. (2023). GPT understands, too. AI Open, 4, 10–20.
8. Zhu, W., & Tan, M. (2023, December). SPT: Learning to selectively insert prompts for better prompt tuning. In Proceedings of the 2023 conference on empirical methods in natural language processing (pp. 11862-11878).
9. Young, S. J. (1996). A review of large-vocabulary continuous-speech recognition. IEEE Signal Processing Magazine, 13(5), 45–57.
10. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780.
11. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., … Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems (pp. 5998–6008).
12. Wolpaw, J. R., & McFarland, D. J. (2004). Control of a two-dimensional movement signal by a noninvasive brain–computer interface in humans. Proceedings of the National Academy of Sciences, 101(51), 17849–17854.
13. Abiri, R., Borhani, S., Sellers, E. W., Jiang, Y., & Zhao, X. (2019). A comprehensive review of EEG-based brain–computer interface paradigms. Journal of Neural Engineering, 16(1), 011001.
14. Huang, J., Patel, A. N., Narasimha, S. M., Mishne, G., & Gilja, V. (2025). Word-Level Error Analysis in Decoding Systems: From Speech Recognition to Brain-Computer Interfaces. In Proc. Interspeech 2025 (pp. 5563-5567).
15. Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in Neural Information Processing Systems (pp. 3104–3112).
16. Graves, A., Mohamed, A., & Hinton, G. (2013). Speech recognition with deep recurrent neural networks. In 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (pp. 6645–6649).
17. Kübler, A., & Birbaumer, N. (2008). Brain–computer interfaces and communication in paralysis: Extinction of goal directed thinking in completely paralysed patients?. Clinical Neurophysiology, 119(11), 2658–2666.
18. Vansteensel, M. J., Pels, E. G., Bleichner, M. G., Branco, M. P., Denison, T., Freudenburg, Z. V., … Ramsey, N. F. (2016). Fully implanted brain–computer interface in a locked-in patient with ALS. New England Journal of Medicine, 375(21), 2060–2066.
19. McMahan, B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. In Artificial Intelligence and Statistics (pp. 1273–1282).
20. Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017). Membership inference attacks against machine learning models. In 2017 IEEE Symposium on Security and Privacy (pp. 3–18).
21. Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., … Amodei, D. (2020). Language models are few-shot learners. In Advances in Neural Information Processing Systems (pp. 1877–1901).
22. Chaudhary, U., Birbaumer, N., & Ramos-Murguialday, A. (2016). Brain–computer interfaces for communication and rehabilitation. Nature Reviews Neurology, 12(9), 513–525.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Engineering Systems and Digital Innovation
This work is licensed under a Creative Commons Attribution 4.0 International License.
