Michael Thornton

Mike investigates how humans make sense of complex signals such as speech. His research explores which features our sensory systems extract in real-world environments, how they remain robust despite the noisiness of everyday life, and how sensory information is combined across different modalities such as vision and hearing. To this end, Mike works with various types of neuroimaging data, for which he also develops specialized signal processing methods.

Mike completed his undergraduate studies in Physics at the University of Oxford, UK, before earning his PhD in neurotechnology under the supervision of Professors Tobias Reichenbach and Danilo Mandic at Imperial College London, UK. Mike spends his free time doing sports, reading, and learning German.

– E. Varano, M. Thornton et al. (2026) “Delta-band cortical speech tracking predicts audiovisual speech-in-noise benefit from natural and simplified visual cues.” NeuroImage, 325, 121654. https://doi.org/10.1088/1741-2552/ac7976.

– M. Thornton et al. (2025) “Comparison of linear and nonlinear methods for decoding selective attention to speech from ear-EEG recordings.” IEEE Access, 13, pp. 127614–127625. https://doi.org/10.1109/ACCESS.2025.3590490.

– M. Thornton et al. (2024) “Detecting gamma-band responses to the speech envelope for the ICASSP 2024 Auditory EEG Decoding Signal Processing Grand Challenge.” Proceedings of ICASSP 2024 Workshops (ICASSPW), pp. 55–56. https://doi.org/10.1109/OJSP.2024.3378593.

– M. Thornton et al. (2024) “Decoding envelope and frequency-following EEG responses to continuous speech using deep neural networks.” IEEE Open Journal of Signal Processing, 5, pp. 700–716. https://doi.org/10.1109/OJSP.2024.3378593.

– M. Yarici et al. (2023) “EEG sensitivity modeling for neural sources and ocular artifacts.” Frontiers in Neuroscience, 16, 997377. https://doi.org/10.3389/fnins.2022.997377.

– M. Thornton et al. (2023) “Relating EEG recordings to speech using envelope tracking and the speech-FFR.” Proceedings of ICASSP 2023. https://doi.org/10.1109/ICASSP49357.2023.10096082.

– M. Thornton et al. (2022) “Robust decoding of the speech envelope from EEG recordings through deep neural networks.” Journal of Neural Engineering, 19(4), 046007. https://doi.org/10.1088/1741-2552/ac7976.