Summary: This perspective review lays out a roadmap for next-generation cognitive brain–computer interfaces (BCIs). To extend BCI applications beyond motor paralysis into psychiatric and cognitive disorders—such as depression, anxiety, PTSD, and OCD—research must move from static motor-cortical decoding to real-time tracking of dynamic, multi-region brain networks.
The review highlights the need to integrate intracranial signal decoding with adaptive neuromodulation to create closed-loop systems that detect emerging psychiatric states and deliver precisely timed neurostimulation responses.
Key Facts
- Motor versus cognitive paradigms: Motor BCIs exploit compact, well-mapped motor cortex representations. Cognitive BCIs must decode widely distributed networks that change rapidly across contexts, where identical neural signals can represent different mental states.
- Combining read and write methods: Cognitive BCIs require uniting two historically separate approaches—neural decoding (reading brain activity) and clinical neuromodulation (stimulating tissue)—into tightly integrated closed-loop systems.
- Real-time detection and response: Next-generation systems aim to recognize dysfunctional cognitive states (for example, depressive crashes, anxiety spikes, or compulsive cycles) as they arise and apply adaptive, time-precise electrical or neurochemical modulation.
- Existing clinical building blocks: High-density intracranial recording, adaptive deep brain stimulation (DBS), and high-resolution neurochemical sensing already exist across clinical and research settings; the challenge is integrating these modalities into compact, low-latency implants.
- Large clinical need and market interest: The population affected by psychiatric and cognitive disorders greatly exceeds that for severe paralysis, attracting growing interest from academic labs, neurotechnology startups, and BCI companies.
Source: Mount Sinai Hospital
Overview: Brain–computer interfaces have transformed restoration of movement and speech for people with paralysis. But the neural basis of cognition—attention, memory, emotion, and decision-making—is fundamentally different from motor control, requiring new concepts and engineering for BCIs that target mental function.
In a perspective published in Trends in Cognitive Sciences, Ignacio Saez, PhD, Director of the Laboratory for Human Neurophysiology at the Icahn School of Medicine at Mount Sinai, proposes a roadmap for developing “cognitive brain–computer interfaces.” The article explains why decoding cognitive states differs from decoding movement, and it outlines the technical advances and organizational collaborations needed to translate these ideas into clinical therapies.
Unlike motor signals, which originate from a relatively compact and stable cortical area, cognitive processes are implemented across broad, interacting networks that reconfigure over time. Because neural representations of attention, memory, and emotion are distributed and context-dependent, simple motor-style decoding algorithms are insufficient; cognitive BCIs must account for dynamic network interactions and changing signal meaning.
Dr. Saez argues that recent advances in intracranial recording, signal decoding, and neuromodulation put the field at an inflection point. By merging robust intracranial monitoring with adaptive stimulation and fast neurochemical sensing, researchers can develop closed-loop systems that monitor internal states and intervene only when clinically necessary, delivering precisely timed, individualized modulation.
The review emphasizes that many foundational tools exist—high-density electrodes, adaptive stimulators, and methods for real-time neurochemical measurement—and that the primary engineering challenge is to integrate multimodal data streams with contextual, low-latency algorithms in implantable form. Success will also require close cooperation between academic researchers, clinicians, regulators, and industry partners to build clinical-grade hardware, software, and regulatory pathways.
Commercial interest is growing: companies that advanced motor and speech BCIs are now exploring cognitive applications because the potential patient population for psychiatric and cognitive conditions is much larger than for paralysis. Translating laboratory advances into approved therapies will depend on validated algorithms, rigorous clinical trials, and multidisciplinary collaboration.
Key Questions Answered:
A: Motor control is localized in well-charted regions like the motor cortex, where neural activity maps directly to physical actions. Cognitive processes are distributed across shifting networks; the same neural pattern can indicate different mental states depending on context. This complexity makes direct application of motor decoding methods ineffective for cognition.
A: Traditional DBS often applies continuous stimulation regardless of brain state. A closed-loop cognitive BCI continuously monitors neural activity, uses adaptive models to detect when a patient is entering a harmful mental state, and delivers calibrated stimulation only when needed, optimizing timing and dosage for effectiveness and safety.
A: Core components include high-density intracranial recording electrodes, adaptive neurostimulators, and fast neurochemical sensing techniques for neurotransmitters such as dopamine and serotonin. The main engineering task is to combine these data streams into a miniaturized, ultra-low-latency implant that runs contextual, adaptive algorithms.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The referenced journal paper was reviewed in full.
- Additional context was added by editorial staff.
About this neurotech and psychology research news
Author: Elizabeth Dowling
Source: Mount Sinai Hospital
Contact: Elizabeth Dowling – Mount Sinai Hospital
Image: The image is credited to Neuroscience News
Original Research: Open access. “The emerging field of cognitive brain–computer interfaces” by Ignacio Saez. Trends in Cognitive Sciences. DOI: 10.1016/j.tics.2026.06.012
Abstract
The emerging field of cognitive brain–computer interfaces
Advances in neural recording and decoding are expanding BCIs beyond motor and language restoration toward monitoring and modulating cognitive functions. Emerging cognitive BCIs aim to track or influence internal states such as attention, memory, emotion, and decision-making. Because cognitive processes are distributed and dynamically reconfigured across brain networks, they pose different challenges than motor or language BCIs and will require new methodologies.
Clinical applications—especially for psychiatric disorders marked by cognitive dysfunction—offer a natural starting point for development. Progress will depend on multimodal datasets, layered behavioral labels, scalable training data, and the integration of foundation-model approaches to build reliable, context-aware decoders. Ultimately, closed-loop devices that combine decoding and neuromodulation are likely to be essential for effective cognitive BCIs, enabling precise, personalized interventions that monitor and shape cognition in real time.
Abstract
Brain–computer interfaces have already restored significant motor and communication function. Extending BCIs to cognition—such as attention and memory—introduces new technical and conceptual challenges because cognitive processes rely on distributed, time-varying neural dynamics rather than localized, stable representations. Neuromodulation provides a complementary path through causal circuit intervention, and integrating decoding with adaptive stimulation in closed-loop systems could bridge systems neuroscience and next-generation neurotherapeutics to monitor and shape human cognition in real time.