Bidirectional brain-computer interface with closed-loop mind-body interactions

The National Center for Complementary and Integrative Health awarded Carnegie Mellon University $616,764 on September 1, 2026, under the Research and Training in Complementary and Integrative Health program (CFDA 93.213) to develop and investigate a bidirectional brain-computer interface (BI-BCI) platform combining electroencephalography-based neural decoding with closed-loop transcranial focused ultrasound neuromodulation.

The award funds two specific research aims. Aim 1 establishes a transcranial focused ultrasound-integrated motion-onset visual evoked potential BCI speller for communication, investigating neural mechanisms of cognitive attention during mind-body interactions by enhancing feature-based attention mechanisms through motion-onset visual stimulation combined with targeted neuromodulation of visual area V5. Aim 2 develops and investigates a transcranial focused ultrasound-modulated sensorimotor rhythm BI-BCI using motor imagery for motor control, testing whether transcranial focused ultrasound modulates event-related sensorimotor dynamics. The dual approach enables real-time measurement of behavior via BCI decoding and adaptive modulation of neural circuits, offering a platform for mechanistic studies of mind-body interventions.

Work is performed in Pittsburgh, Pennsylvania. The period of performance extends through June 30, 2031.

Abstract

Brain-computer interfaces (BCIS) enable direct communication between the brain and the external world. We propose to establish and investigate the mechanisms of a novel, noninvasive bidirectional BCI (BI-BCI) platform. This system combines electroencephalography (EEG)-based neural decoding ("reading" brain activity) with closed-loop, low-intensity transcranial focused ultrasound (TFUS) neuromodulation ("writing" to the brain). This dual approach enables real-time measurement of behavior via BCI decoding and adaptive modulation of neural circuits. By integrating noninvasive neural decoding with precision neuromodulation, the proposed BI-BCI offers a transformative platform for mechanistic studies of mind-body interventions-revealing how specific brain activity patterns relate to behavior and how targeted modulation can shift these patterns to achieve desired outcomes. To realize this vision, we propose two specific aims unified under a central theme: to develop and investigate a BI-BCI framework that engages both the "mind" (intention and cognition) and "body" (physiological processes) for two key applications-communication prostheses (aim 1) and motor prostheses (aim 2). This foundational work will catalyze a new class of mind-body interventions. Aim 1: establish a TFUS-integrated motion-onset visual evoked potential BCI speller for communication and investigate neural mechanisms of cognitive attention during mind-body interactions. We will enhance feature-based attention mechanisms by combining motion-onset visual stimulation with targeted neuromodulation of visual area V5. This approach will yield a novel BI-BCI with significantly improved accuracy and information throughput. Aim 2: establish and investigate a TFUS-modulated sensorimotor rhythm (SMR) BI-BCI using motor imagery for motor control. We will test the hypothesis that TFUS modulates the dynamics of event-related desynchronization/synchronization (ERD/ERS), thereby enhancing motor imagery-based BCI control of an external object. Successful completion of this research will significantly enhance BCI performance and yield critical insights into the neural mechanisms underlying mind-body interventions made possible by neuromodulation-enhanced, closed-loop BI-BCIS.