A different approach to digital mental health.
80% of mental health app users stop within two weeks. Text-heavy tools demand cognitive effort during distress. AI chatbots fail at crisis detection.
We build something different: physiological, non-verbal tools that work through the body, not around it. Research-backed. Open-source. Designed to complement existing platforms like mindLAMP, not compete with them.
Why Physiological? Why Non-Verbal?
These aren't aesthetic choices. They're responses to documented problems.
Larger Effect Sizes
HRV biofeedback shows effect sizes of d = 0.81 for anxiety. Digital CBT shows d = 0.4-0.6. Physiological approaches consistently outperform purely cognitive digital interventions.
Goessl et al., 2017; Cuijpers et al., 2019
Works During Distress
Text-heavy apps require reading and cognitive effort—exactly what's hardest during anxiety or depression. Visual breathing pacers and biophilic patterns work when words can't reach you.
No literacy barrier. No language barrier.
Integration, Not Competition
Platforms like mindLAMP already do measurement and phenotyping well. We build open tools designed to complement that ecosystem—usable standalone or as integrated activities.
Building on Dr. Torous's work, not duplicating it.
What's Different Here
Each design choice addresses a documented failure in existing approaches.
| Problem with Prior Approaches | This Approach | Evidence |
|---|---|---|
| Text-heavy apps require cognitive effort during distress | Non-verbal physiological tools — visual pacers, biophilic patterns | HRV d=0.81 vs CBT d=0.54 |
| AI chatbots fail at crisis detection (78% miss rate) | No chatbots — human escalation always ≤2 taps away | Brown University, 2025 |
| Apps lose 80% of users in 2 weeks | Minimal friction design — no streaks, no guilt, no accounts required | Baumel et al., 2019 |
| Algorithmic personalization creates filter bubbles and privacy risks | User-controlled preferences — on-device, transparent, exportable | Privacy-preserving by design |
| Siloed tools duplicate existing research infrastructure | Integration-first — designed to complement mindLAMP, not compete | Open-source, documented APIs |
Two Research Tracks
Individual tools and systems reform. Both matter.
Individual Digital Tools
What works, what doesn't, and why physiological approaches outperform text-heavy apps by nearly 2x on effect size.
Systems Reform
160K+ provider shortage. 23+ hour ED boarding. How technology can support—not substitute for—structural change.
Start Here
Read the Whitepaper
Complete synthesis: evidence on digital tools, systems reform data, the framework, contributed tools. 80+ pages with citations.
Try the Tools
Resonance breathing, biophilic visuals, observer cultivation. Works in 3-5 minutes. No account needed.
For Developers
AI safety framework, design patterns, anti-patterns, mindLAMP integration. What the evidence says to build.
Also: Research • For Clinicians • For Institutions • Ethics • Attribution
What We Won't Do
Boundaries matter. Here are ours.
- We won't claim therapeutic benefit without evidence. If something might help, we say "might." If evidence is limited, we say so.
- We won't pretend AI can replace human connection. The research is clear: hybrid approaches work best.
- We won't hide our sources. Everything is cited. You can check our work.
- We won't promise what technology can't deliver. Digital tools have real limitations. We discuss them.
- We won't ignore safety. Crisis protocols, escalation paths, and human oversight are non-negotiable.
Part of Something Larger
MindBridge is part of the 22b1 Research Initiative, which works on cross-system medicine, AI fairness, and global health. Mental health is health. These issues connect.
Visit 22b1.com →This is a research site, not a crisis service. If you or someone you know is in immediate danger:
- 988 Suicide & Crisis Lifeline: Call or text 988 (US)
- Crisis Text Line: Text HOME to 741741
- Emergency: Call 911 or go to your nearest emergency department
- International: Find a helpline in your country