Training methods where both user strategies and decoder parameters adapt continuously.

Central store for source code and related files, often integrated with collaboration and CI/CD tools.

A set of tests covering multiple cognitive domains (e.g., memory, attention, executive function).

A structured psychotherapy focused on modifying unhelpful thoughts and behaviors to improve mood and functioning.

The capacity to switch between tasks, strategies, or mental sets in response to changing demands.

The total mental effort required to perform a task or handle information at a given moment.

Restructuring unhelpful thoughts to reduce distress or improve functioning.

Ability to maintain cognitive performance despite stress, fatigue, or age-related changes.

Inferring internal states such as workload, attention, or intention from neural signals.

Patterns of thinking that make a person more susceptible to depression or anxiety.

Framework estimating mental effort from task metrics or physiology.

BCIs estimating mental effort from neural signals to adapt interfaces or tasks.

The set of actions available through a BCI (e.g., left, right, select).

Referencing scheme that subtracts the average of all channels from each channel to reduce shared noise.

Measure of how well an amplifier suppresses noise common to all input channels.

Repetitive, difficult-to-control behaviors driven by urges rather than goals.

Mathematical model describing neural dynamics or network behavior.

X-ray-based imaging that reconstructs cross-sectional brain images, often used in acute clinical settings.

Gel applied between electrodes and skin to reduce impedance and improve current delivery.

A CT variant using a cone-shaped X-ray beam, often applied in head and neck or interventional contexts.

A comprehensive map of structural or functional connections between brain regions.

Selecting stimulation sites based on network connectivity rather than only anatomy.

The electrical resistance at the electrode–tissue interface, affecting stimulation efficiency.

Packaging code and dependencies into isolated units (e.g., Docker containers) for reproducible deployments.

Decoding that integrates neural data with task or environmental context to improve accuracy.

TBS pattern typically associated with net inhibitory effects on cortical excitability.

Automated pipelines that test, build, and deploy code or models when changes are made.

A neural feature mapped to a specific BCI command or output.

The ease with which cortical neurons can be activated, often modulated by stimulation protocols.

The distance between the white and pial surfaces of cortex, often used as a marker of structural integrity.

Low-intensity electrical stimulation applied via ear clips or scalp to modulate mood or sleep.

Interaction between neural oscillations at different frequencies, sometimes used as a control feature in BCIs.

Technique splitting data into subsets to estimate model generalizability.

Unwanted interference between recording channels or nearby electrodes.

Shaping current flow through multiple electrodes to focus or shift stimulation effects.

Systems enabling cursor movement through decoded neural activity.

Enhancing device security to prevent breaches, tampering, or unauthorized data access.

Visual interface that aggregates metrics, trends, and alerts for monitoring systems or cohorts.

Hardware-software stack that records biosignals (e.g., EEG, fNIRS, EMG) and streams them to a computer in real time.

Standardized rules for labeling signals, images, or behaviors consistently.

Structured reference defining variables, units, formats, and permissible values.

Policies and processes that define how data is collected, stored, accessed, and used responsibly.

Processes that standardize datasets collected across devices, protocols, or study sites.

Software that receives, validates, and routes incoming data streams from devices or APIs.

Large storage repository that holds raw, heterogeneous data (signals, logs, metadata) at scale.

Governance of data from collection through storage to deletion.

End-to-end process that moves data from acquisition through cleaning, transformation, and storage.

Documentation of where data came from and how it was processed across its lifecycle.

The process of replacing personal identifiers with coded substitutes so data can be linked without revealing identity, while still allowing re-identification under controlled conditions.

Automated or manual tests verifying completeness, consistency, and validity of data.

Centralized, structured database optimized for analytics and reporting on curated datasets.

FDA pathway for first-of-kind devices with no predicate but low-to-moderate risk.

Broad process of removing or generalizing personal identifiers to reduce privacy risk.

Collecting training data to optimize BCI decoding performance.

Gradual changes in neural signal properties that reduce decoding accuracy unless updated.

Using machine learning to predict stimuli, tasks, or states from brain activity patterns.

TMS using specialized coils (e.g., H-coils) designed to reach deeper brain structures.

Invasive neuromodulation using implanted electrodes in deep brain structures, typically driven by a subcutaneous stimulator.

A large-scale brain network active at rest and deactivated during many goal-directed tasks.

Slow brain oscillations typically around 0.5–4 Hz, prominent in deep sleep and some pathological states.

Fixed, false belief not aligned with cultural norms and resistant to contradictory evidence.

A syndrome of substantial cognitive decline that interferes with daily functioning, often progressive.

The computational and physical setting where a system runs (clinic, cloud, OR, wearable).

Reduced responsiveness to repeated stimulation, requiring parameter adjustment.

Signal distortion introduced by hardware or electrical interference.

Procedure ensuring a device’s measurements remain accurate.

Embedded software running directly on neurotech hardware.

Coordinated oversight of versioning, maintenance, updates, and retirement of devices.

Optical imaging method reconstructing 3D maps of cortical hemodynamics using near-infrared light.

A diffusion MRI method that models directional water diffusion to infer white matter pathways.

MRI technique measuring the diffusion of water molecules, sensitive to microstructural changes and acute stroke.

A quantifiable behavioral or physiological signal measured by digital tools that relates to health or disease.

Infrastructure that ingests sensor data, extracts features, and computes clinically relevant metrics.

Behavioral or sensor-derived metric that reflects cognitive function in everyday environments.

A clinically meaningful metric derived from sensor data in trials or remote monitoring.

Technology-enabled healthcare including apps, sensors, telehealth, and analytics.

System extracting behavioral and physiological markers from continuous sensor data.

Software-based intervention with demonstrated clinical benefit, often delivered via apps or online platforms.

Computational model simulating a person’s neural system for prediction or personalization.

A stimulation lead with segmented contacts that steer current toward specific tissue regions.

Infrastructure enabling parallel processing across multiple machines or clusters.

ML methods aligning data distributions across subjects, sessions, or devices to improve generalization.

Stimulation of dorsal root ganglia to modulate sensory pathways and chronic pain.

The combination of parameters (intensity, frequency, duration, number of sessions) defining a stimulation regimen.

Systematic adjustment of stimulation or therapy parameters to maximize benefit.

Relationship between stimulation intensity/duration and physiological or behavioral effects.

A study design in which neither participants nor investigators know which intervention each participant receives, reducing expectancy and observer bias.

Adjustment applied when sensors or baselines shift slowly over time.

The proportion of time stimulation is “on” versus “off” within a given period.

A state of unease or dissatisfaction, often used to describe negative mood.

Extract-Transform-Load (or Load-Transform) workflow that ingests raw data and prepares it for analysis.

Fast MRI acquisition method commonly used in fMRI and diffusion imaging.

The time between excitation and signal readout, influencing MRI contrast.

Repeated, in-the-moment data collection in real-world settings via phones or wearables.

Running neural network inference directly on devices (e.g., wearables, implants) instead of cloud.

Running inference or preprocessing on-device or near the sensor, rather than in the cloud.

Cryptographically protected firmware on clinical devices to prevent tampering and unauthorized updates.

Workflow distributing processing between local devices and cloud systems.

Directional influence one brain region exerts over another, often inferred with causal models.

Measurement of cardiac electrical activity, often synchronized with neural recordings.