A deep learning approach for training attention with a neuromorphic brain-computerinterface (BCI)

Faculty: Prof. Sridharan Devarajan (CNS & CSA) and Prof. Chetan Singh Thakur (DESE)

Description: In this project we will attempt to answer the following question: Can we train
people to improve their ability to pay attention using signals measured from their own brain?

To answer this question we will develop a brain computer interface (BCI) by recording brain
signals with high-resolution functional magnetic resonance imaging (fMRI) and
electroencephalography (EEG) when human participants perform attention tasks.

We will develop and refine state-of-the-art deep learning models, include transformers and
state-space models, to decode where and to what object the person attending to, using a
gamified Android interface. Based on this decoding analysis, we will provide neuro-feedback
to subjects to make them aware of their level of attention, and potential lapses of attention to
help them improve their performance. The entire system including the recording, analysis and
neurofeedback will be implemented on a BCI neuromorphic chip, which allows efficient, lowpower
decoding and real-time feedback.

The project lies at the intersection of neuroscience, computer science, electronic systems
and deep learning, and is part of the Brain Co-processors Neural Implants project at IISc.

Link: https://brain-computation.iisc.ac.in/moonshot-project/