How Sway Improves Focus, Clarity, and Cognitive Vitality
Our smart glasses use real-time brain and biometric signals to help you enter flow, recover faster, and build lasting mental strength.
Leveraging Decades of Peer-Reviewed Research in Neuroscience, Psychology, and Physiology
We hypothesize that continuous, multimodal monitoring of brain and physiological signals - combined with individualized baselines and temporal trends - can enable accurate modeling of cognitive states across varying contexts, ultimately supporting improvements in focus, cognitive recovery, and mental resilience. Our hypothesis is supported by 6 key pillars of research.
At the foundation of Sway’s system are three validated physiological signals: EEG, EOG, and PPG.
When combined, these signals enable rich insight into mental states across time - from momentary focus to long-term resilience.
Flow is a highly efficient mental state marked by deep concentration and reduced self-monitoring. Decades of research link flow to increased alpha and theta power in specific brain regions, along with physiological signs of calm arousal.
By tracking neural oscillations and heart dynamics in real time, we detect biomarkers that correspond to task engagement, distraction, and overload. This enables the modeling of transient cognitive states, providing a foundation for adaptive feedback and long-term skill building.
Cognitive recovery is not passive - it’s rhythmic. Neural and autonomic systems follow daily cycles and react to workload, stress, and mental fatigue.
By analyzing markers like alpha frequency band dynamics, HRV patterns, and eye movement changes, we identify whether your brain is overworked, under-recovered, or entering restorative states. These markers let us surface recovery needs before they manifest as burnout.
Short-term states matter - but long-term baselines tell the full story. Changes in cognitive performance, stress sensitivity, and recovery efficiency unfold gradually and can be tracked using slow-moving metrics.
We establish rolling baselines for peak alpha frequency, HRV, and key EEG ratios to detect deviations from your norm. This allows us to detect early signs of strain, growth, or imbalance - enabling timely intervention and longitudinal feedback.
Each signal alone provides a piece of the picture. Together, they allow us to resolve ambiguity and improve accuracy.
EEG may show a dip in attention, but EOG tells us if it’s due to drowsiness, distraction, or gaze aversion. PPG confirms if sympathetic activation aligns with the cognitive shift. This multimodal context improves classification performance and builds robustness to noise and motion artifacts.
Real-world data is messy. We integrate modern signal processing and machine learning techniques to clean, interpret, and classify mental states.
Artifact removal (e.g. ICA, ASR), time-frequency analysis, and lightweight edge models allow us to extract features in both streaming and passive modes. With supervised models trained on labeled data and semi-supervised learning for personalization, we ensure accurate tracking of states like focus, fatigue, and cognitive readiness - even during all-day wear.